Do you take any of the following actions when using Gen AI tools? Check all that apply

What does your organization have?

What types of challenges have you or your team faced in regards to using gen AI responsibly?

Q1: For what types of work tasks/purposes do you use Gen AI (check all that apply)?

(worktasks <- genaiclean %>%
  select(ai_usage_prod:ai_usage_other) %>%
   ungroup() %>%
  dplyr::summarize_all(sum) %>%
   t()) %>%
  as.data.frame() %>%
    rename(Count = 1)

workresources <- c(
  "As a component of a product we are developing (e.g., using/integrating gen AI in new products)",
  "Content generation (e.g., written content to support marketing and customer engagement)",
  "Productivity tasks (e.g., summarizing meeting notes, writing emails)",
  "UI/UX design",
  "Product ideation and brainstorming",
  "Data analysis and insights",
  "Coding",
  "Customer support automation",
  "Other"
)

rownames(worktasks) <- workresources
toString(genaiclean %>% as.data.frame() %>% filter(Q1_4_TEXT != "") %>% select(Q1_4_TEXT))

## [1] "c(\"To eliminate the blank page in product descriptions, brochures, and scripts\", \"Proofreading the packaging info.\")"

What gen AI tools/models do you use?

(genaitools <- genaiclean %>%
  select(foundmodels:externalopensource) %>%
   ungroup() %>%
  dplyr::summarize_all(sum) %>%
   t() %>%
  as.data.frame() %>%
    rename(Count = 1))

genaitoolsname = c(
   "My company's own gen AI tools/models developed by our research teams using our own foundational model(s)",
   "My company's own gen AI tools/models developed by our research teams based on third-party foundation model(s)",
   "External proprietary gen AI model",
   "External open-source gen AI tool")

rownames(genaitools) <- genaitoolsname
paste("Those who listed more than one tool:", genaiclean %>% filter(numtools > 1) %>% nrow())

## [1] "Those who listed more than one tool: 67"

% of women

formattable::percent(table(genaiclean$gender)/nrow(genaiclean))

## 
##    Female      Male NonBinary 
##    46.41%    53.04%     0.55%

Race distribution

Note: had trouble with race variable in R. Qualtrics downloads the data so that each cell can have multiple values (i.e., 1,7), which

formattable::percent(table(genaiclean$race)/nrow(genaiclean))

## 
## American Indian           Black      East Asian     Multiracial           Other     South Asian           White 
##           0.55%          18.23%           3.31%           3.31%           5.52%           7.18%          61.88%

Age Distribution

formattable::percent(table(genaiclean$age)/nrow(genaiclean))

## 
##   18-24   25-34   35-44   45-54   55-64 65 over 
##    8.3%   40.9%   28.2%   14.9%    6.1%    1.7%

Country Distribution

formattable::percent(table(genaiclean$D1)/nrow(genaiclean))

## 
##                   Canada                   CANADA                  Denmark                  England                 England                   Germany                    India                   India                   Ireland                 New york                 Scotland                    Spain                     U.S.                       uk                       Uk                       UK           United Kingdom           UNITED KINGDOM          united Kingdom           United Kingdom   United state of america  United State Of America  UNITED STATE OF AMERICA            united states            United states            United States           United States            UNITED STATES  united states of america United States of America                       us                       US                      usa                      Usa                      USA                     USA  
##                    7.73%                    0.55%                    0.55%                    1.66%                    1.10%                    2.21%                    1.10%                    0.55%                    0.55%                    0.55%                    0.55%                    0.55%                    0.55%                    1.10%                    0.55%                    5.52%                   11.05%                    1.10%                    0.55%                    4.97%                    0.55%                    0.55%                    0.55%                    1.10%                    1.66%                   23.20%                    2.76%                    0.55%                    1.10%                    5.52%                    0.55%                    3.31%                    2.21%                    0.55%                   12.15%                    0.55%

Roles

formattable::percent(table(genaiclean$role)/nrow(genaiclean))

## 
##          Engineer  Product designer   Product Manager Product marketing 
##              7.7%              7.2%             47.5%             27.6%

Number of employees

formattable::percent(table(genaiclean$approxnum)/nrow(genaiclean))

## 
##      1. 1-49    2. 50-499  3. 499-1000 4. 1000-4999     5. 5000+ 
##        26.5%        33.1%        14.4%         9.4%        14.4%

Power hierarchy

paste("Mean role power hierarchy:", mean(genaiclean$OD2)); paste("Median role power hierarchy:", median(genaiclean$OD2)); paste("SD role power hierarchy:", sd(genaiclean$OD2))

## [1] "Mean role power hierarchy: 5.35911602209945"

## [1] "Median role power hierarchy: 6"

## [1] "SD role power hierarchy: 2.19805148378125"

Industry

Printed off the text of people’s responses to “other” category

formattable::percent(table(genaiclean$industry)/nrow(genaiclean))

## 
##                    Agr. and farm                  Arts and enter.                     Construction Customer Relationship Management                        Education            Finance and insuranec            Healthcare and pharm.                      Hospitality               IT and/or software                    Manufacturing                           Retail               Telecommunications     Transportation and logistics 
##                            2.76%                            6.08%                            2.76%                            2.21%                            3.31%                           12.15%                            5.52%                            2.76%                           19.34%                           17.13%                           11.05%                            0.55%                            5.52%

toString(genaiclean %>% filter(OD4_14_TEXT != "") %>% select(OD4_14_TEXT) )

## [1] "c(\"Nonprofit human services \", \"MARKETING COORDINATOR \", \"Meteorology and science\", \"Online Retail \", \"Wholesale Trade\", \"Management\", \"consulting agency\", \"Safety and Consulting\", \"Engineering\", \"Publishing \", \"consultant\", \"Non-profit\", \"Charity \", \"We resell vintage clothing, accessories, collectibles, and home decor\", \"Professional services\", \"Government & Public Administration\")"

Predictions 1 - Firm Size and Adoption of Open Vs. Closed Models

Prediction A. Smaller firms are more likely to adopt open models because of cost
Prediction B. Larger firms are more likely to use open models because they can easily retrain on their own data

adoptionopenclosed <- genaiclean %>%
  select(approxnum, foundmodels:externalopensource) %>% as.data.frame() 

for(column in 2:ncol(adoptionopenclosed)){
  print(genaitoolsname[column-1])
  print(table(cbind(as.data.frame(adoptionopenclosed["approxnum"]), as.data.frame(adoptionopenclosed[column]))))
  print(chisq.test(table(cbind(as.data.frame(adoptionopenclosed["approxnum"]), as.data.frame(adoptionopenclosed[column])))))
  y <- colnames(adoptionopenclosed[column])
  yvariable <- as.numeric(unlist(as.data.frame(adoptionopenclosed[y])))
  print(summary(glm(yvariable~approxnum, adoptionopenclosed, family = "binomial")))
  print(summary(glm(yvariable~approxnum, adoptionopenclosed, family = "binomial"))$coefficient)
  left_join( 
  as.data.frame(exp(summary(glm(yvariable~approxnum, adoptionopenclosed, family = "binomial"))$coefficients))["Estimate"] %>% rownames_to_column(),
  exp(confint(glm(yvariable~approxnum, adoptionopenclosed, family = "binomial"))) %>% as.data.frame() %>% rownames_to_column())
}

## [1] "My company's own gen AI tools/models developed by our research teams using our own foundational model(s)"
##               foundmodels
## approxnum       0  1
##   1. 1-49      41  7
##   2. 50-499    46 14
##   3. 499-1000  20  6
##   4. 1000-4999 12  5
##   5. 5000+     20  6

## Warning in chisq.test(table(cbind(as.data.frame(adoptionopenclosed["approxnum"]), : Chi-squared approximation may be incorrect

## 
##  Pearson's Chi-squared test
## 
## data:  table(cbind(as.data.frame(adoptionopenclosed["approxnum"]), as.data.frame(adoptionopenclosed[column])))
## X-squared = 2.2, df = 4, p-value = 0.7
## 
## 
## Call:
## glm(formula = yvariable ~ approxnum, family = "binomial", data = adoptionopenclosed)
## 
## Coefficients:
##                       Estimate Std. Error z value Pr(>|z|)    
## (Intercept)             -1.768      0.409   -4.32 0.000015 ***
## approxnum2. 50-499       0.578      0.510    1.13     0.26    
## approxnum3. 499-1000     0.564      0.620    0.91     0.36    
## approxnum4. 1000-4999    0.892      0.671    1.33     0.18    
## approxnum5. 5000+        0.564      0.620    0.91     0.36    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 184.12  on 176  degrees of freedom
## Residual deviance: 181.85  on 172  degrees of freedom
##   (4 observations deleted due to missingness)
## AIC: 191.9
## 
## Number of Fisher Scoring iterations: 4
## 
##                       Estimate Std. Error z value   Pr(>|z|)
## (Intercept)            -1.7677     0.4090 -4.3223 0.00001544
## approxnum2. 50-499      0.5781     0.5103  1.1328 0.25729808
## approxnum3. 499-1000    0.5637     0.6196  0.9098 0.36295352
## approxnum4. 1000-4999   0.8922     0.6713  1.3291 0.18379974
## approxnum5. 5000+       0.5637     0.6196  0.9098 0.36295352

## Waiting for profiling to be done...

## Joining with `by = join_by(rowname)`

## [1] "My company's own gen AI tools/models developed by our research teams based on third-party foundation model(s)"
##               thirdparty
## approxnum       0  1
##   1. 1-49      45  3
##   2. 50-499    46 14
##   3. 499-1000  18  8
##   4. 1000-4999 12  5
##   5. 5000+     18  8

## Warning in chisq.test(table(cbind(as.data.frame(adoptionopenclosed["approxnum"]), : Chi-squared approximation may be incorrect

## 
##  Pearson's Chi-squared test
## 
## data:  table(cbind(as.data.frame(adoptionopenclosed["approxnum"]), as.data.frame(adoptionopenclosed[column])))
## X-squared = 10, df = 4, p-value = 0.04
## 
## 
## Call:
## glm(formula = yvariable ~ approxnum, family = "binomial", data = adoptionopenclosed)
## 
## Coefficients:
##                       Estimate Std. Error z value  Pr(>|z|)    
## (Intercept)             -2.708      0.596   -4.54 0.0000056 ***
## approxnum2. 50-499       1.518      0.670    2.27    0.0234 *  
## approxnum3. 499-1000     1.897      0.732    2.59    0.0096 ** 
## approxnum4. 1000-4999    1.833      0.799    2.29    0.0219 *  
## approxnum5. 5000+        1.897      0.732    2.59    0.0096 ** 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 184.12  on 176  degrees of freedom
## Residual deviance: 172.43  on 172  degrees of freedom
##   (4 observations deleted due to missingness)
## AIC: 182.4
## 
## Number of Fisher Scoring iterations: 5
## 
##                       Estimate Std. Error z value    Pr(>|z|)
## (Intercept)             -2.708     0.5963  -4.542 0.000005584
## approxnum2. 50-499       1.518     0.6699   2.267 0.023400893
## approxnum3. 499-1000     1.897     0.7322   2.591 0.009569439
## approxnum4. 1000-4999    1.833     0.7993   2.293 0.021863770
## approxnum5. 5000+        1.897     0.7322   2.591 0.009569439

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "External proprietary gen AI model"
##               externalprop
## approxnum       0  1
##   1. 1-49       8 40
##   2. 50-499    12 48
##   3. 499-1000   4 22
##   4. 1000-4999  1 16
##   5. 5000+      4 22

## Warning in chisq.test(table(cbind(as.data.frame(adoptionopenclosed["approxnum"]), : Chi-squared approximation may be incorrect

## 
##  Pearson's Chi-squared test
## 
## data:  table(cbind(as.data.frame(adoptionopenclosed["approxnum"]), as.data.frame(adoptionopenclosed[column])))
## X-squared = 2, df = 4, p-value = 0.7
## 
## 
## Call:
## glm(formula = yvariable ~ approxnum, family = "binomial", data = adoptionopenclosed)
## 
## Coefficients:
##                       Estimate Std. Error z value Pr(>|z|)    
## (Intercept)             1.6094     0.3873    4.16 0.000032 ***
## approxnum2. 50-499     -0.2231     0.5041   -0.44     0.66    
## approxnum3. 499-1000    0.0953     0.6674    0.14     0.89    
## approxnum4. 1000-4999   1.1632     1.1011    1.06     0.29    
## approxnum5. 5000+       0.0953     0.6674    0.14     0.89    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 157.88  on 176  degrees of freedom
## Residual deviance: 155.56  on 172  degrees of freedom
##   (4 observations deleted due to missingness)
## AIC: 165.6
## 
## Number of Fisher Scoring iterations: 5
## 
##                       Estimate Std. Error z value   Pr(>|z|)
## (Intercept)            1.60944     0.3873  4.1556 0.00003245
## approxnum2. 50-499    -0.22314     0.5041 -0.4426 0.65804503
## approxnum3. 499-1000   0.09531     0.6674  0.1428 0.88644570
## approxnum4. 1000-4999  1.16315     1.1011  1.0563 0.29081733
## approxnum5. 5000+      0.09531     0.6674  0.1428 0.88644570

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "External open-source gen AI tool"
##               externalopensource
## approxnum       0  1
##   1. 1-49      35 13
##   2. 50-499    53  7
##   3. 499-1000  19  7
##   4. 1000-4999 13  4
##   5. 5000+     21  5

## Warning in chisq.test(table(cbind(as.data.frame(adoptionopenclosed["approxnum"]), : Chi-squared approximation may be incorrect

## 
##  Pearson's Chi-squared test
## 
## data:  table(cbind(as.data.frame(adoptionopenclosed["approxnum"]), as.data.frame(adoptionopenclosed[column])))
## X-squared = 5, df = 4, p-value = 0.3
## 
## 
## Call:
## glm(formula = yvariable ~ approxnum, family = "binomial", data = adoptionopenclosed)
## 
## Coefficients:
##                       Estimate Std. Error z value Pr(>|z|)   
## (Intercept)           -0.99040    0.32480   -3.05   0.0023 **
## approxnum2. 50-499    -1.03398    0.51693   -2.00   0.0455 * 
## approxnum3. 499-1000  -0.00813    0.54862   -0.01   0.9882   
## approxnum4. 1000-4999 -0.18826    0.65758   -0.29   0.7747   
## approxnum5. 5000+     -0.44469    0.59423   -0.75   0.4543   
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 178.79  on 176  degrees of freedom
## Residual deviance: 173.60  on 172  degrees of freedom
##   (4 observations deleted due to missingness)
## AIC: 183.6
## 
## Number of Fisher Scoring iterations: 4
## 
##                       Estimate Std. Error  z value Pr(>|z|)
## (Intercept)           -0.99040     0.3248 -3.04927 0.002294
## approxnum2. 50-499    -1.03398     0.5169 -2.00024 0.045474
## approxnum3. 499-1000  -0.00813     0.5486 -0.01482 0.988176
## approxnum4. 1000-4999 -0.18826     0.6576 -0.28628 0.774660
## approxnum5. 5000+     -0.44469     0.5942 -0.74834 0.454258

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

Prediction 2:

genderadoption <- genaiclean %>%
  select(gender, ai_usage_prod:ai_usage_other) %>% as.data.frame() %>%
  filter(gender != "NonBinary")

for(column in 2:ncol(genderadoption)){
  print(workresources[column-1])
  print(table(cbind(as.data.frame(genderadoption["gender"]), as.data.frame(genderadoption[column]))))
  print(chisq.test(table(cbind(as.data.frame(genderadoption["gender"]), as.data.frame(genderadoption[column])))))    
}

## [1] "As a component of a product we are developing (e.g., using/integrating gen AI in new products)"
##         ai_usage_prod
## gender    0  1
##   Female 12 72
##   Male   26 70
## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(genderadoption["gender"]), as.data.frame(genderadoption[column])))
## X-squared = 3.7, df = 1, p-value = 0.06
## 
## [1] "Content generation (e.g., written content to support marketing and customer engagement)"
##         ai_usage_content
## gender    0  1
##   Female 24 60
##   Male   40 56
## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(genderadoption["gender"]), as.data.frame(genderadoption[column])))
## X-squared = 2.8, df = 1, p-value = 0.09
## 
## [1] "Productivity tasks (e.g., summarizing meeting notes, writing emails)"
##         ai_usage_prodtasks
## gender    0  1
##   Female 25 59
##   Male   38 58
## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(genderadoption["gender"]), as.data.frame(genderadoption[column])))
## X-squared = 1.5, df = 1, p-value = 0.2
## 
## [1] "UI/UX design"
##         ai_usage_uidesign
## gender    0  1
##   Female 75  9
##   Male   83 13
## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(genderadoption["gender"]), as.data.frame(genderadoption[column])))
## X-squared = 0.12, df = 1, p-value = 0.7
## 
## [1] "Product ideation and brainstorming"
##         ai_usage_productideation
## gender    0  1
##   Female 41 43
##   Male   56 40
## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(genderadoption["gender"]), as.data.frame(genderadoption[column])))
## X-squared = 1.3, df = 1, p-value = 0.3
## 
## [1] "Data analysis and insights"
##         ai_usage_data
## gender    0  1
##   Female 50 34
##   Male   49 47
## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(genderadoption["gender"]), as.data.frame(genderadoption[column])))
## X-squared = 0.98, df = 1, p-value = 0.3
## 
## [1] "Coding"
##         ai_usage_coding
## gender    0  1
##   Female 72 12
##   Male   77 19
## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(genderadoption["gender"]), as.data.frame(genderadoption[column])))
## X-squared = 0.61, df = 1, p-value = 0.4
## 
## [1] "Customer support automation"
##         ai_usage_customer
## gender    0  1
##   Female 58 26
##   Male   75 21
## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(genderadoption["gender"]), as.data.frame(genderadoption[column])))
## X-squared = 1.5, df = 1, p-value = 0.2
## 
## [1] "Other"
##         ai_usage_other
## gender    0  1
##   Female 83  1
##   Male   95  1

## Warning in chisq.test(table(cbind(as.data.frame(genderadoption["gender"]), : Chi-squared approximation may be incorrect

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(genderadoption["gender"]), as.data.frame(genderadoption[column])))
## X-squared = 0.000000000000000000000000000029, df = 1, p-value = 1

adoption <- genaiclean %>%
  select(gender, takeaction_workwithcol:takeaction_noactions) %>% as.data.frame() %>%
  filter(gender != "NonBinary")

for(column in 2:ncol(adoption)){
  print(rowsaction[column-1])
  print(table(cbind(as.data.frame(adoption["gender"]), as.data.frame(adoption[column]))))
  print(chisq.test(table(cbind(as.data.frame(adoption["gender"]), as.data.frame(adoption[column])))))    
}

## [1] "Work with colleauges who have job responsibility"
##         takeaction_workwithcol
## gender    0  1
##   Female 49 35
##   Male   47 49
## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(adoption["gender"]), as.data.frame(adoption[column])))
## X-squared = 1.2, df = 1, p-value = 0.3
## 
## [1] "Take ethical AI trainings"
##         takeaction_ethicaltraining
## gender    0  1
##   Female 47 37
##   Male   50 46
## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(adoption["gender"]), as.data.frame(adoption[column])))
## X-squared = 0.14, df = 1, p-value = 0.7
## 
## [1] "Conduct audits of the gen ai tools"
##         takeaction_audits
## gender    0  1
##   Female 61 23
##   Male   65 31
## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(adoption["gender"]), as.data.frame(adoption[column])))
## X-squared = 0.31, df = 1, p-value = 0.6
## 
## [1] "Conduct fairness and bias testing"
##         takeaction_testing
## gender    0  1
##   Female 58 26
##   Male   61 35
## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(adoption["gender"]), as.data.frame(adoption[column])))
## X-squared = 0.39, df = 1, p-value = 0.5
## 
## [1] "Consider data privacy implications"
##         takeaction_privacy
## gender    0  1
##   Female 49 35
##   Male   52 44
## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(adoption["gender"]), as.data.frame(adoption[column])))
## X-squared = 0.17, df = 1, p-value = 0.7
## 
## [1] "Ask about the data or model to understand potential limitations"
##         takeaction_askaboutdata
## gender    0  1
##   Female 51 33
##   Male   59 37
## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(adoption["gender"]), as.data.frame(adoption[column])))
## X-squared = 0, df = 1, p-value = 1
## 
## [1] "Conduct adversarial testing"
##         takeaction_conductadversarial
## gender    0  1
##   Female 71 13
##   Male   85 11
## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(adoption["gender"]), as.data.frame(adoption[column])))
## X-squared = 0.33, df = 1, p-value = 0.6
## 
## [1] "Use explainability methods"
##         takeaction_explainability
## gender    0  1
##   Female 58 26
##   Male   63 33
## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(adoption["gender"]), as.data.frame(adoption[column])))
## X-squared = 0.11, df = 1, p-value = 0.7
## 
## [1] "Build transparency approaches"
##         takeaction_transparencyapproaches
## gender    0  1
##   Female 61 23
##   Male   66 30
## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(adoption["gender"]), as.data.frame(adoption[column])))
## X-squared = 0.16, df = 1, p-value = 0.7
## 
## [1] "No actions taken"
##         takeaction_noactions
## gender    0  1
##   Female 67 17
##   Male   76 20
## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(adoption["gender"]), as.data.frame(adoption[column])))
## X-squared = 0, df = 1, p-value = 1

Prediction 3: Firm size predict Adoption of responsible AI strategies

responsiblegenai <- genaiclean %>%
  select(approxnum, takeaction_workwithcol:takeaction_noactions) %>% as.data.frame() 

adoptionname <- c(
  "Work with colleagues who have job responsibilities",
  "Take ethical AI trainings",
  "Conduct audits of the gen AI tools",
  "Conduct fairness testing",
  "Consider data privacy implications and take actions to protect data privacy",
  "Ask about the data or model to understand potential limitations it has",
  "Conduct adversarial testing or red teaming",
  "Use explainability methods",
  "Build transparency approaches",
  "No actions taken"
)

for(column in 2:ncol(responsiblegenai)){
  print(adoptionname[column-1])
  print(
    graphtable <- as.data.frame(table(cbind(as.data.frame(responsiblegenai["approxnum"]), as.data.frame(responsiblegenai[column])))) %>%
  rename(yesno = 2) %>%
  mutate(yesno = case_when(yesno == 0 ~ "No",
                           yesno == 1 ~ "Yes")))
  print(graphtable %>%
    ggplot(aes(x = approxnum, y = Freq, fill = yesno)) +
    geom_bar(stat = "identity", position = "dodge")+
    labs(
      x = "Company Size",
      y = "Count"
      )+
    jtools::theme_apa())
  
  print(chisq.test(table(cbind(as.data.frame(responsiblegenai["approxnum"]), as.data.frame(responsiblegenai[column])))))
  y <- colnames(responsiblegenai[column])
  yvariable <- as.numeric(unlist(as.data.frame(responsiblegenai[y])))
  print(summary(glm(yvariable~approxnum, responsiblegenai, family = "binomial")))
  print(summary(glm(yvariable~approxnum, responsiblegenai, family = "binomial"))$coefficient)
  left_join( 
  as.data.frame(exp(summary(glm(yvariable~approxnum, responsiblegenai, family = "binomial"))$coefficients))["Estimate"] %>% rownames_to_column(),
  exp(confint(glm(yvariable~approxnum, responsiblegenai, family = "binomial"))) %>% as.data.frame() %>% rownames_to_column())
}

## [1] "Work with colleagues who have job responsibilities"
##       approxnum yesno Freq
## 1       1. 1-49    No   34
## 2     2. 50-499    No   29
## 3   3. 499-1000    No   12
## 4  4. 1000-4999    No    8
## 5      5. 5000+    No   13
## 6       1. 1-49   Yes   14
## 7     2. 50-499   Yes   31
## 8   3. 499-1000   Yes   14
## 9  4. 1000-4999   Yes    9
## 10     5. 5000+   Yes   13

## 
##  Pearson's Chi-squared test
## 
## data:  table(cbind(as.data.frame(responsiblegenai["approxnum"]), as.data.frame(responsiblegenai[column])))
## X-squared = 7.4, df = 4, p-value = 0.1
## 
## 
## Call:
## glm(formula = yvariable ~ approxnum, family = "binomial", data = responsiblegenai)
## 
## Coefficients:
##                       Estimate Std. Error z value Pr(>|z|)   
## (Intercept)             -0.887      0.318   -2.79   0.0052 **
## approxnum2. 50-499       0.954      0.409    2.33   0.0198 * 
## approxnum3. 499-1000     1.041      0.506    2.06   0.0394 * 
## approxnum4. 1000-4999    1.005      0.580    1.73   0.0834 . 
## approxnum5. 5000+        0.887      0.505    1.76   0.0787 . 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 244.1  on 176  degrees of freedom
## Residual deviance: 236.5  on 172  degrees of freedom
##   (4 observations deleted due to missingness)
## AIC: 246.5
## 
## Number of Fisher Scoring iterations: 4
## 
##                       Estimate Std. Error z value Pr(>|z|)
## (Intercept)            -0.8873     0.3176  -2.794 0.005203
## approxnum2. 50-499      0.9540     0.4094   2.330 0.019784
## approxnum3. 499-1000    1.0415     0.5056   2.060 0.039403
## approxnum4. 1000-4999   1.0051     0.5805   1.731 0.083365
## approxnum5. 5000+       0.8873     0.5047   1.758 0.078713

## Waiting for profiling to be done...

## Joining with `by = join_by(rowname)`

## [1] "Take ethical AI trainings"
##       approxnum yesno Freq
## 1       1. 1-49    No   36
## 2     2. 50-499    No   31
## 3   3. 499-1000    No    6
## 4  4. 1000-4999    No    9
## 5      5. 5000+    No   14
## 6       1. 1-49   Yes   12
## 7     2. 50-499   Yes   29
## 8   3. 499-1000   Yes   20
## 9  4. 1000-4999   Yes    8
## 10     5. 5000+   Yes   12

## 
##  Pearson's Chi-squared test
## 
## data:  table(cbind(as.data.frame(responsiblegenai["approxnum"]), as.data.frame(responsiblegenai[column])))
## X-squared = 19, df = 4, p-value = 0.0009
## 
## 
## Call:
## glm(formula = yvariable ~ approxnum, family = "binomial", data = responsiblegenai)
## 
## Coefficients:
##                       Estimate Std. Error z value Pr(>|z|)    
## (Intercept)             -1.099      0.333   -3.30  0.00098 ***
## approxnum2. 50-499       1.032      0.422    2.45  0.01441 *  
## approxnum3. 499-1000     2.303      0.573    4.02 0.000058 ***
## approxnum4. 1000-4999    0.981      0.589    1.66  0.09601 .  
## approxnum5. 5000+        0.944      0.516    1.83  0.06700 .  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 244.10  on 176  degrees of freedom
## Residual deviance: 224.58  on 172  degrees of freedom
##   (4 observations deleted due to missingness)
## AIC: 234.6
## 
## Number of Fisher Scoring iterations: 4
## 
##                       Estimate Std. Error z value   Pr(>|z|)
## (Intercept)            -1.0986     0.3333  -3.296 0.00098128
## approxnum2. 50-499      1.0319     0.4217   2.447 0.01440883
## approxnum3. 499-1000    2.3026     0.5725   4.022 0.00005774
## approxnum4. 1000-4999   0.9808     0.5893   1.665 0.09600798
## approxnum5. 5000+       0.9445     0.5156   1.832 0.06700054

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "Conduct audits of the gen AI tools"
##       approxnum yesno Freq
## 1       1. 1-49    No   41
## 2     2. 50-499    No   39
## 3   3. 499-1000    No   15
## 4  4. 1000-4999    No   11
## 5      5. 5000+    No   18
## 6       1. 1-49   Yes    7
## 7     2. 50-499   Yes   21
## 8   3. 499-1000   Yes   11
## 9  4. 1000-4999   Yes    6
## 10     5. 5000+   Yes    8

## 
##  Pearson's Chi-squared test
## 
## data:  table(cbind(as.data.frame(responsiblegenai["approxnum"]), as.data.frame(responsiblegenai[column])))
## X-squared = 8.3, df = 4, p-value = 0.08
## 
## 
## Call:
## glm(formula = yvariable ~ approxnum, family = "binomial", data = responsiblegenai)
## 
## Coefficients:
##                       Estimate Std. Error z value Pr(>|z|)    
## (Intercept)             -1.768      0.409   -4.32 0.000015 ***
## approxnum2. 50-499       1.149      0.490    2.34    0.019 *  
## approxnum3. 499-1000     1.458      0.570    2.56    0.011 *  
## approxnum4. 1000-4999    1.162      0.652    1.78    0.075 .  
## approxnum5. 5000+        0.957      0.590    1.62    0.105    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 216.08  on 176  degrees of freedom
## Residual deviance: 207.17  on 172  degrees of freedom
##   (4 observations deleted due to missingness)
## AIC: 217.2
## 
## Number of Fisher Scoring iterations: 4
## 
##                       Estimate Std. Error z value   Pr(>|z|)
## (Intercept)            -1.7677     0.4090  -4.322 0.00001544
## approxnum2. 50-499      1.1486     0.4904   2.342 0.01917344
## approxnum3. 499-1000    1.4575     0.5699   2.557 0.01054784
## approxnum4. 1000-4999   1.1615     0.6518   1.782 0.07473771
## approxnum5. 5000+       0.9567     0.5897   1.622 0.10474515

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "Conduct fairness testing"
##       approxnum yesno Freq
## 1       1. 1-49    No   34
## 2     2. 50-499    No   40
## 3   3. 499-1000    No   13
## 4  4. 1000-4999    No   10
## 5      5. 5000+    No   20
## 6       1. 1-49   Yes   14
## 7     2. 50-499   Yes   20
## 8   3. 499-1000   Yes   13
## 9  4. 1000-4999   Yes    7
## 10     5. 5000+   Yes    6

## 
##  Pearson's Chi-squared test
## 
## data:  table(cbind(as.data.frame(responsiblegenai["approxnum"]), as.data.frame(responsiblegenai[column])))
## X-squared = 5.3, df = 4, p-value = 0.3
## 
## 
## Call:
## glm(formula = yvariable ~ approxnum, family = "binomial", data = responsiblegenai)
## 
## Coefficients:
##                       Estimate Std. Error z value Pr(>|z|)   
## (Intercept)             -0.887      0.318   -2.79   0.0052 **
## approxnum2. 50-499       0.194      0.419    0.46   0.6434   
## approxnum3. 499-1000     0.887      0.505    1.76   0.0787 . 
## approxnum4. 1000-4999    0.531      0.586    0.91   0.3654   
## approxnum5. 5000+       -0.317      0.563   -0.56   0.5741   
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 226.69  on 176  degrees of freedom
## Residual deviance: 221.50  on 172  degrees of freedom
##   (4 observations deleted due to missingness)
## AIC: 231.5
## 
## Number of Fisher Scoring iterations: 4
## 
##                       Estimate Std. Error z value Pr(>|z|)
## (Intercept)            -0.8873     0.3176 -2.7942 0.005203
## approxnum2. 50-499      0.1942     0.4193  0.4630 0.643356
## approxnum3. 499-1000    0.8873     0.5047  1.7582 0.078713
## approxnum4. 1000-4999   0.5306     0.5863  0.9051 0.365406
## approxnum5. 5000+      -0.3167     0.5635 -0.5620 0.574122

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "Consider data privacy implications and take actions to protect data privacy"
##       approxnum yesno Freq
## 1       1. 1-49    No   29
## 2     2. 50-499    No   34
## 3   3. 499-1000    No    9
## 4  4. 1000-4999    No   12
## 5      5. 5000+    No   17
## 6       1. 1-49   Yes   19
## 7     2. 50-499   Yes   26
## 8   3. 499-1000   Yes   17
## 9  4. 1000-4999   Yes    5
## 10     5. 5000+   Yes    9

## 
##  Pearson's Chi-squared test
## 
## data:  table(cbind(as.data.frame(responsiblegenai["approxnum"]), as.data.frame(responsiblegenai[column])))
## X-squared = 7.6, df = 4, p-value = 0.1
## 
## 
## Call:
## glm(formula = yvariable ~ approxnum, family = "binomial", data = responsiblegenai)
## 
## Coefficients:
##                       Estimate Std. Error z value Pr(>|z|)  
## (Intercept)             -0.423      0.295   -1.43    0.152  
## approxnum2. 50-499       0.155      0.394    0.39    0.695  
## approxnum3. 499-1000     1.059      0.507    2.09    0.037 *
## approxnum4. 1000-4999   -0.453      0.609   -0.74    0.457  
## approxnum5. 5000+       -0.213      0.507   -0.42    0.674  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 241.83  on 176  degrees of freedom
## Residual deviance: 234.23  on 172  degrees of freedom
##   (4 observations deleted due to missingness)
## AIC: 244.2
## 
## Number of Fisher Scoring iterations: 4
## 
##                       Estimate Std. Error z value Pr(>|z|)
## (Intercept)            -0.4229     0.2952 -1.4327  0.15195
## approxnum2. 50-499      0.1546     0.3937  0.3927  0.69455
## approxnum3. 499-1000    1.0588     0.5070  2.0885  0.03676
## approxnum4. 1000-4999  -0.4526     0.6086 -0.7436  0.45709
## approxnum5. 5000+      -0.2131     0.5070 -0.4204  0.67421

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "Ask about the data or model to understand potential limitations it has"
##       approxnum yesno Freq
## 1       1. 1-49    No   33
## 2     2. 50-499    No   36
## 3   3. 499-1000    No   16
## 4  4. 1000-4999    No    8
## 5      5. 5000+    No   17
## 6       1. 1-49   Yes   15
## 7     2. 50-499   Yes   24
## 8   3. 499-1000   Yes   10
## 9  4. 1000-4999   Yes    9
## 10     5. 5000+   Yes    9

## 
##  Pearson's Chi-squared test
## 
## data:  table(cbind(as.data.frame(responsiblegenai["approxnum"]), as.data.frame(responsiblegenai[column])))
## X-squared = 2.8, df = 4, p-value = 0.6
## 
## 
## Call:
## glm(formula = yvariable ~ approxnum, family = "binomial", data = responsiblegenai)
## 
## Coefficients:
##                       Estimate Std. Error z value Pr(>|z|)  
## (Intercept)             -0.788      0.311   -2.53    0.011 *
## approxnum2. 50-499       0.383      0.408    0.94    0.348  
## approxnum3. 499-1000     0.318      0.509    0.63    0.532  
## approxnum4. 1000-4999    0.906      0.577    1.57    0.116  
## approxnum5. 5000+        0.152      0.517    0.30    0.768  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 234.82  on 176  degrees of freedom
## Residual deviance: 232.08  on 172  degrees of freedom
##   (4 observations deleted due to missingness)
## AIC: 242.1
## 
## Number of Fisher Scoring iterations: 4
## 
##                       Estimate Std. Error z value Pr(>|z|)
## (Intercept)            -0.7885     0.3114 -2.5320  0.01134
## approxnum2. 50-499      0.3830     0.4079  0.9388  0.34781
## approxnum3. 499-1000    0.3185     0.5094  0.6252  0.53185
## approxnum4. 1000-4999   0.9062     0.5771  1.5702  0.11636
## approxnum5. 5000+       0.1525     0.5166  0.2951  0.76790

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "Conduct adversarial testing or red teaming"
##       approxnum yesno Freq
## 1       1. 1-49    No   43
## 2     2. 50-499    No   57
## 3   3. 499-1000    No   18
## 4  4. 1000-4999    No   13
## 5      5. 5000+    No   23
## 6       1. 1-49   Yes    5
## 7     2. 50-499   Yes    3
## 8   3. 499-1000   Yes    8
## 9  4. 1000-4999   Yes    4
## 10     5. 5000+   Yes    3

## Warning in chisq.test(table(cbind(as.data.frame(responsiblegenai["approxnum"]), : Chi-squared approximation may be incorrect

## 
##  Pearson's Chi-squared test
## 
## data:  table(cbind(as.data.frame(responsiblegenai["approxnum"]), as.data.frame(responsiblegenai[column])))
## X-squared = 13, df = 4, p-value = 0.01
## 
## 
## Call:
## glm(formula = yvariable ~ approxnum, family = "binomial", data = responsiblegenai)
## 
## Coefficients:
##                       Estimate Std. Error z value  Pr(>|z|)    
## (Intercept)             -2.152      0.472   -4.55 0.0000053 ***
## approxnum2. 50-499      -0.793      0.758   -1.05     0.295    
## approxnum3. 499-1000     1.341      0.635    2.11     0.035 *  
## approxnum4. 1000-4999    0.973      0.742    1.31     0.190    
## approxnum5. 5000+        0.115      0.775    0.15     0.882    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 136.74  on 176  degrees of freedom
## Residual deviance: 125.14  on 172  degrees of freedom
##   (4 observations deleted due to missingness)
## AIC: 135.1
## 
## Number of Fisher Scoring iterations: 5
## 
##                       Estimate Std. Error z value    Pr(>|z|)
## (Intercept)            -2.1518     0.4725 -4.5540 0.000005264
## approxnum2. 50-499     -0.7927     0.7577 -1.0462 0.295481124
## approxnum3. 499-1000    1.3408     0.6355  2.1100 0.034857253
## approxnum4. 1000-4999   0.9731     0.7417  1.3119 0.189545727
## approxnum5. 5000+       0.1149     0.7746  0.1483 0.882104857

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "Use explainability methods"
##       approxnum yesno Freq
## 1       1. 1-49    No   35
## 2     2. 50-499    No   40
## 3   3. 499-1000    No   12
## 4  4. 1000-4999    No   12
## 5      5. 5000+    No   21
## 6       1. 1-49   Yes   13
## 7     2. 50-499   Yes   20
## 8   3. 499-1000   Yes   14
## 9  4. 1000-4999   Yes    5
## 10     5. 5000+   Yes    5

## 
##  Pearson's Chi-squared test
## 
## data:  table(cbind(as.data.frame(responsiblegenai["approxnum"]), as.data.frame(responsiblegenai[column])))
## X-squared = 8.3, df = 4, p-value = 0.08
## 
## 
## Call:
## glm(formula = yvariable ~ approxnum, family = "binomial", data = responsiblegenai)
## 
## Coefficients:
##                       Estimate Std. Error z value Pr(>|z|)   
## (Intercept)             -0.990      0.325   -3.05   0.0023 **
## approxnum2. 50-499       0.297      0.425    0.70   0.4841   
## approxnum3. 499-1000     1.145      0.510    2.24   0.0249 * 
## approxnum4. 1000-4999    0.115      0.624    0.18   0.8538   
## approxnum5. 5000+       -0.445      0.594   -0.75   0.4543   
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 222.45  on 176  degrees of freedom
## Residual deviance: 214.40  on 172  degrees of freedom
##   (4 observations deleted due to missingness)
## AIC: 224.4
## 
## Number of Fisher Scoring iterations: 4
## 
##                       Estimate Std. Error z value Pr(>|z|)
## (Intercept)            -0.9904     0.3248 -3.0493 0.002294
## approxnum2. 50-499      0.2973     0.4248  0.6997 0.484134
## approxnum3. 499-1000    1.1445     0.5102  2.2435 0.024862
## approxnum4. 1000-4999   0.1149     0.6236  0.1843 0.853768
## approxnum5. 5000+      -0.4447     0.5942 -0.7483 0.454258

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "Build transparency approaches"
##       approxnum yesno Freq
## 1       1. 1-49    No   40
## 2     2. 50-499    No   41
## 3   3. 499-1000    No   16
## 4  4. 1000-4999    No   10
## 5      5. 5000+    No   20
## 6       1. 1-49   Yes    8
## 7     2. 50-499   Yes   19
## 8   3. 499-1000   Yes   10
## 9  4. 1000-4999   Yes    7
## 10     5. 5000+   Yes    6

## Warning in chisq.test(table(cbind(as.data.frame(responsiblegenai["approxnum"]), : Chi-squared approximation may be incorrect

## 
##  Pearson's Chi-squared test
## 
## data:  table(cbind(as.data.frame(responsiblegenai["approxnum"]), as.data.frame(responsiblegenai[column])))
## X-squared = 6.6, df = 4, p-value = 0.2
## 
## 
## Call:
## glm(formula = yvariable ~ approxnum, family = "binomial", data = responsiblegenai)
## 
## Coefficients:
##                       Estimate Std. Error z value Pr(>|z|)    
## (Intercept)             -1.609      0.387   -4.16 0.000032 ***
## approxnum2. 50-499       0.840      0.476    1.76    0.078 .  
## approxnum3. 499-1000     1.139      0.559    2.04    0.042 *  
## approxnum4. 1000-4999    1.253      0.627    2.00    0.046 *  
## approxnum5. 5000+        0.405      0.606    0.67    0.503    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 210.73  on 176  degrees of freedom
## Residual deviance: 203.95  on 172  degrees of freedom
##   (4 observations deleted due to missingness)
## AIC: 213.9
## 
## Number of Fisher Scoring iterations: 4
## 
##                       Estimate Std. Error z value   Pr(>|z|)
## (Intercept)            -1.6094     0.3873 -4.1556 0.00003245
## approxnum2. 50-499      0.8403     0.4765  1.7636 0.07779729
## approxnum3. 499-1000    1.1394     0.5590  2.0383 0.04152174
## approxnum4. 1000-4999   1.2528     0.6268  1.9987 0.04563885
## approxnum5. 5000+       0.4055     0.6055  0.6696 0.50311034

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "No actions taken"
##       approxnum yesno Freq
## 1       1. 1-49    No   32
## 2     2. 50-499    No   50
## 3   3. 499-1000    No   25
## 4  4. 1000-4999    No   11
## 5      5. 5000+    No   22
## 6       1. 1-49   Yes   16
## 7     2. 50-499   Yes   10
## 8   3. 499-1000   Yes    1
## 9  4. 1000-4999   Yes    6
## 10     5. 5000+   Yes    4

## Warning in chisq.test(table(cbind(as.data.frame(responsiblegenai["approxnum"]), : Chi-squared approximation may be incorrect

## 
##  Pearson's Chi-squared test
## 
## data:  table(cbind(as.data.frame(responsiblegenai["approxnum"]), as.data.frame(responsiblegenai[column])))
## X-squared = 12, df = 4, p-value = 0.02
## 
## 
## Call:
## glm(formula = yvariable ~ approxnum, family = "binomial", data = responsiblegenai)
## 
## Coefficients:
##                       Estimate Std. Error z value Pr(>|z|)  
## (Intercept)             -0.693      0.306   -2.26    0.024 *
## approxnum2. 50-499      -0.916      0.462   -1.98    0.047 *
## approxnum3. 499-1000    -2.526      1.064   -2.37    0.018 *
## approxnum4. 1000-4999    0.087      0.593    0.15    0.883  
## approxnum5. 5000+       -1.012      0.624   -1.62    0.105  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 181.49  on 176  degrees of freedom
## Residual deviance: 168.05  on 172  degrees of freedom
##   (4 observations deleted due to missingness)
## AIC: 178
## 
## Number of Fisher Scoring iterations: 5
## 
##                       Estimate Std. Error z value Pr(>|z|)
## (Intercept)           -0.69315     0.3062 -2.2638  0.02359
## approxnum2. 50-499    -0.91629     0.4623 -1.9819  0.04749
## approxnum3. 499-1000  -2.52573     1.0644 -2.3728  0.01765
## approxnum4. 1000-4999  0.08701     0.5927  0.1468  0.88329
## approxnum5. 5000+     -1.01160     0.6239 -1.6215  0.10491

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

Prediction 4:

orgsresponsiblegenai <- genaiclean %>%
  select(orghave_leadership:orghave_neither,
         takeaction_workwithcol:takeaction_noactions) %>% as.data.frame() %>%
  mutate_at(
    vars(orghave_leadership:orghave_neither),
    funs(
      case_when(
        . == 0 ~ "No",
        . == 1 ~ "Yes")))

## Warning: `funs()` was deprecated in dplyr 0.8.0.
## ℹ Please use a list of either functions or lambdas:
## 
## # Simple named list: list(mean = mean, median = median)
## 
## # Auto named with `tibble::lst()`: tibble::lst(mean, median)
## 
## # Using lambdas list(~ mean(., trim = .2), ~ median(., na.rm = TRUE))
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was generated.

ethicalpolicies <- c(
  "Leadership that has expressed commitment to responsible AI",
  "Responsible or ethical AI principles",
  "Policies that inform the use of gen AI",
  "Clear incentives for using implementing gen AI responsibly",
  "Neither"
)

adoptionname <- c(
  "Work with colleagues who have job responsibilities",
  "Take ethical AI trainings",
  "Conduct audits of the gen AI tools",
  "Conduct fairness testing",
  "Consider data privacy implications and take actions to protect data privacy",
  "Ask about the data or model to understand potential limitations it has",
  "Conduct adversarial testing or red teaming",
  "Use explainability methods",
  "Build transparency approaches",
  "No actions taken"
)

for(iv in 1:5){
  for(column in 6:ncol(orgsresponsiblegenai)){
    print(colnames(orgsresponsiblegenai)[iv])
    print(adoptionname[column-1])
    print(
      graphtable <- as.data.frame(table(cbind(as.data.frame(orgsresponsiblegenai[iv]),
                                              as.data.frame(orgsresponsiblegenai[column]))))%>%
        rename(variable = 1,
               yesno = 2) %>%
        mutate(yesno = case_when(yesno == 0 ~ "No",
                                 yesno == 1 ~ "Yes")))
  print(
    graphtable %>%
      ggplot(aes(x = variable, y = Freq, fill = yesno)) +
      geom_bar(stat = "identity", position = "dodge")+
      labs(
        x = colnames(orgsresponsiblegenai)[iv],
        y = "Count")+
      jtools::theme_apa())
  
  print(chisq.test(table(cbind(as.data.frame(orgsresponsiblegenai[iv]),
                               as.data.frame(orgsresponsiblegenai[column])))))
  
  independentvariable <- as.character(unlist(orgsresponsiblegenai[iv]))
  
  y <- colnames(orgsresponsiblegenai[column])
  
  yvariable <- as.numeric(unlist(as.data.frame(orgsresponsiblegenai[y])))
  
  print(summary(glm(yvariable~independentvariable, orgsresponsiblegenai, family = "binomial")))
  
  print(summary(glm(yvariable~independentvariable, orgsresponsiblegenai, family = "binomial"))$coefficient)
  
  left_join(
    as.data.frame(exp(summary(glm(yvariable~independentvariable, orgsresponsiblegenai, family = "binomial"))$coefficients))["Estimate"] %>% 
      rownames_to_column(),
  exp(confint(glm(yvariable~independentvariable, orgsresponsiblegenai, family = "binomial"))) %>% as.data.frame() %>% rownames_to_column())
}}

## [1] "orghave_leadership"
## [1] "Consider data privacy implications and take actions to protect data privacy"
##   variable yesno Freq
## 1       No    No   69
## 2      Yes    No   28
## 3       No   Yes   34
## 4      Yes   Yes   50

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 16, df = 1, p-value = 0.00006
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value Pr(>|z|)    
## (Intercept)              -0.708      0.210   -3.38  0.00073 ***
## independentvariableYes    1.288      0.316    4.08 0.000045 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 249.98  on 180  degrees of freedom
## Residual deviance: 232.50  on 179  degrees of freedom
## AIC: 236.5
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value   Pr(>|z|)
## (Intercept)             -0.7077     0.2095  -3.378 0.00073091
## independentvariableYes   1.2876     0.3156   4.079 0.00004515

## Waiting for profiling to be done...

## Joining with `by = join_by(rowname)`

## [1] "orghave_leadership"
## [1] "Ask about the data or model to understand potential limitations it has"
##   variable yesno Freq
## 1       No    No   72
## 2      Yes    No   25
## 3       No   Yes   31
## 4      Yes   Yes   53

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 24, df = 1, p-value = 0.0000009
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value   Pr(>|z|)    
## (Intercept)              -0.843      0.215   -3.92 0.00008754 ***
## independentvariableYes    1.594      0.324    4.92 0.00000087 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 249.98  on 180  degrees of freedom
## Residual deviance: 223.86  on 179  degrees of freedom
## AIC: 227.9
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value     Pr(>|z|)
## (Intercept)             -0.8427     0.2148  -3.923 0.0000875434
## independentvariableYes   1.5941     0.3241   4.919 0.0000008693

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_leadership"
## [1] "Conduct adversarial testing or red teaming"
##   variable yesno Freq
## 1       No    No   83
## 2      Yes    No   43
## 3       No   Yes   20
## 4      Yes   Yes   35

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 12, df = 1, p-value = 0.0004
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value    Pr(>|z|)    
## (Intercept)              -1.423      0.249   -5.71 0.000000011 ***
## independentvariableYes    1.217      0.337    3.61     0.00031 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 222.31  on 180  degrees of freedom
## Residual deviance: 208.71  on 179  degrees of freedom
## AIC: 212.7
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value      Pr(>|z|)
## (Intercept)              -1.423     0.2491  -5.713 0.00000001109
## independentvariableYes    1.217     0.3375   3.607 0.00030955063

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_leadership"
## [1] "Use explainability methods"
##   variable yesno Freq
## 1       No    No   78
## 2      Yes    No   42
## 3       No   Yes   25
## 4      Yes   Yes   36

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 8.6, df = 1, p-value = 0.003
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value   Pr(>|z|)    
## (Intercept)              -1.138      0.230   -4.95 0.00000074 ***
## independentvariableYes    0.984      0.323    3.04     0.0023 ** 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 231.33  on 180  degrees of freedom
## Residual deviance: 221.83  on 179  degrees of freedom
## AIC: 225.8
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value    Pr(>|z|)
## (Intercept)             -1.1378     0.2298  -4.951 0.000000739
## independentvariableYes   0.9837     0.3231   3.044 0.002332165

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_leadership"
## [1] "Build transparency approaches"
##   variable yesno Freq
## 1       No    No   64
## 2      Yes    No   37
## 3       No   Yes   39
## 4      Yes   Yes   41

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 3.3, df = 1, p-value = 0.07
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value Pr(>|z|)  
## (Intercept)              -0.495      0.203   -2.44    0.015 *
## independentvariableYes    0.598      0.304    1.96    0.050 *
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 248.48  on 180  degrees of freedom
## Residual deviance: 244.59  on 179  degrees of freedom
## AIC: 248.6
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value Pr(>|z|)
## (Intercept)             -0.4953     0.2031  -2.438  0.01476
## independentvariableYes   0.5980     0.3044   1.964  0.04951

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_leadership"
## [1] "No actions taken"
##   variable yesno Freq
## 1       No    No   70
## 2      Yes    No   41
## 3       No   Yes   33
## 4      Yes   Yes   37

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 3.8, df = 1, p-value = 0.05
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value Pr(>|z|)    
## (Intercept)              -0.752      0.211   -3.56  0.00037 ***
## independentvariableYes    0.649      0.310    2.10  0.03611 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 241.55  on 180  degrees of freedom
## Residual deviance: 237.12  on 179  degrees of freedom
## AIC: 241.1
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value  Pr(>|z|)
## (Intercept)             -0.7520     0.2112  -3.561 0.0003691
## independentvariableYes   0.6493     0.3098   2.096 0.0361133

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_leadership"
## [1] NA
##   variable yesno Freq
## 1       No    No   94
## 2      Yes    No   63
## 3       No   Yes    9
## 4      Yes   Yes   15

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 3.4, df = 1, p-value = 0.07
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value       Pr(>|z|)    
## (Intercept)              -2.346      0.349   -6.72 0.000000000018 ***
## independentvariableYes    0.911      0.452    2.02          0.044 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 141.65  on 180  degrees of freedom
## Residual deviance: 137.43  on 179  degrees of freedom
## AIC: 141.4
## 
## Number of Fisher Scoring iterations: 5
## 
##                        Estimate Std. Error z value         Pr(>|z|)
## (Intercept)              -2.346     0.3489  -6.724 0.00000000001772
## independentvariableYes    0.911     0.4520   2.016 0.04384928390533

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_leadership"
## [1] NA
##   variable yesno Freq
## 1       No    No   76
## 2      Yes    No   45
## 3       No   Yes   27
## 4      Yes   Yes   33

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 4.5, df = 1, p-value = 0.03
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value  Pr(>|z|)    
## (Intercept)              -1.035      0.224   -4.62 0.0000039 ***
## independentvariableYes    0.725      0.320    2.26     0.024 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 229.95  on 180  degrees of freedom
## Residual deviance: 224.78  on 179  degrees of freedom
## AIC: 228.8
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value    Pr(>|z|)
## (Intercept)             -1.0349     0.2240  -4.619 0.000003852
## independentvariableYes   0.7247     0.3205   2.261 0.023741570

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_leadership"
## [1] NA
##   variable yesno Freq
## 1       No    No   86
## 2      Yes    No   41
## 3       No   Yes   17
## 4      Yes   Yes   37

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 19, df = 1, p-value = 0.00001
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value    Pr(>|z|)    
## (Intercept)              -1.621      0.265   -6.11 0.000000001 ***
## independentvariableYes    1.518      0.349    4.35 0.000013629 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 220.62  on 180  degrees of freedom
## Residual deviance: 200.20  on 179  degrees of freedom
## AIC: 204.2
## 
## Number of Fisher Scoring iterations: 3
## 
##                        Estimate Std. Error z value       Pr(>|z|)
## (Intercept)              -1.621     0.2654  -6.108 0.000000001011
## independentvariableYes    1.518     0.3491   4.350 0.000013628964

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_leadership"
## [1] NA
##   variable yesno Freq
## 1       No    No   72
## 2      Yes    No   72
## 3       No   Yes   31
## 4      Yes   Yes    6

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 12, df = 1, p-value = 0.0004
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value Pr(>|z|)    
## (Intercept)              -0.843      0.215   -3.92 0.000088 ***
## independentvariableYes   -1.642      0.476   -3.45  0.00056 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 183.34  on 180  degrees of freedom
## Residual deviance: 168.31  on 179  degrees of freedom
## AIC: 172.3
## 
## Number of Fisher Scoring iterations: 5
## 
##                        Estimate Std. Error z value   Pr(>|z|)
## (Intercept)             -0.8427     0.2148  -3.923 0.00008754
## independentvariableYes  -1.6422     0.4761  -3.449 0.00056246

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_responsible"
## [1] "Consider data privacy implications and take actions to protect data privacy"
##   variable yesno Freq
## 1       No    No   64
## 2      Yes    No   33
## 3       No   Yes   34
## 4      Yes   Yes   50

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 11, df = 1, p-value = 0.001
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value Pr(>|z|)    
## (Intercept)              -0.633      0.212   -2.98  0.00288 ** 
## independentvariableYes    1.048      0.309    3.39  0.00069 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 249.98  on 180  degrees of freedom
## Residual deviance: 238.08  on 179  degrees of freedom
## AIC: 242.1
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value  Pr(>|z|)
## (Intercept)             -0.6325     0.2122  -2.981 0.0028776
## independentvariableYes   1.0480     0.3088   3.394 0.0006882

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_responsible"
## [1] "Ask about the data or model to understand potential limitations it has"
##   variable yesno Freq
## 1       No    No   75
## 2      Yes    No   22
## 3       No   Yes   23
## 4      Yes   Yes   61

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 43, df = 1, p-value = 0.00000000005
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value      Pr(>|z|)    
## (Intercept)              -1.182      0.238   -4.96 0.00000070845 ***
## independentvariableYes    2.202      0.344    6.39 0.00000000016 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 249.98  on 180  degrees of freedom
## Residual deviance: 202.79  on 179  degrees of freedom
## AIC: 206.8
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value        Pr(>|z|)
## (Intercept)              -1.182     0.2384  -4.959 0.0000007084457
## independentvariableYes    2.202     0.3445   6.392 0.0000000001638

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_responsible"
## [1] "Conduct adversarial testing or red teaming"
##   variable yesno Freq
## 1       No    No   76
## 2      Yes    No   50
## 3       No   Yes   22
## 4      Yes   Yes   33

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 5.6, df = 1, p-value = 0.02
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value  Pr(>|z|)    
## (Intercept)              -1.240      0.242   -5.12 0.0000003 ***
## independentvariableYes    0.824      0.330    2.50     0.013 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 222.31  on 180  degrees of freedom
## Residual deviance: 215.93  on 179  degrees of freedom
## AIC: 219.9
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value     Pr(>|z|)
## (Intercept)             -1.2397     0.2421  -5.121 0.0000003046
## independentvariableYes   0.8242     0.3300   2.497 0.0125134264

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_responsible"
## [1] "Use explainability methods"
##   variable yesno Freq
## 1       No    No   80
## 2      Yes    No   40
## 3       No   Yes   18
## 4      Yes   Yes   43

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 21, df = 1, p-value = 0.000005
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value    Pr(>|z|)    
## (Intercept)              -1.492      0.261   -5.72 0.000000011 ***
## independentvariableYes    1.564      0.341    4.59 0.000004522 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 231.33  on 180  degrees of freedom
## Residual deviance: 208.43  on 179  degrees of freedom
## AIC: 212.4
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value      Pr(>|z|)
## (Intercept)              -1.492     0.2609  -5.718 0.00000001078
## independentvariableYes    1.564     0.3410   4.586 0.00000452165

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_responsible"
## [1] "Build transparency approaches"
##   variable yesno Freq
## 1       No    No   58
## 2      Yes    No   43
## 3       No   Yes   40
## 4      Yes   Yes   40

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 0.71, df = 1, p-value = 0.4
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value Pr(>|z|)  
## (Intercept)              -0.372      0.206   -1.81    0.071 .
## independentvariableYes    0.299      0.301    0.99    0.320  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 248.48  on 180  degrees of freedom
## Residual deviance: 247.49  on 179  degrees of freedom
## AIC: 251.5
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value Pr(>|z|)
## (Intercept)             -0.3716     0.2055 -1.8079  0.07063
## independentvariableYes   0.2992     0.3008  0.9947  0.31987

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_responsible"
## [1] "No actions taken"
##   variable yesno Freq
## 1       No    No   67
## 2      Yes    No   44
## 3       No   Yes   31
## 4      Yes   Yes   39

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 3.8, df = 1, p-value = 0.05
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value Pr(>|z|)    
## (Intercept)              -0.771      0.217   -3.55  0.00039 ***
## independentvariableYes    0.650      0.309    2.10  0.03546 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 241.55  on 180  degrees of freedom
## Residual deviance: 237.08  on 179  degrees of freedom
## AIC: 241.1
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value Pr(>|z|)
## (Intercept)             -0.7707     0.2172  -3.548 0.000388
## independentvariableYes   0.6501     0.3091   2.103 0.035463

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_responsible"
## [1] NA
##   variable yesno Freq
## 1       No    No   93
## 2      Yes    No   64
## 3       No   Yes    5
## 4      Yes   Yes   19

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 11, df = 1, p-value = 0.001
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value      Pr(>|z|)    
## (Intercept)              -2.923      0.459   -6.37 0.00000000019 ***
## independentvariableYes    1.709      0.528    3.24        0.0012 ** 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 141.65  on 180  degrees of freedom
## Residual deviance: 128.80  on 179  degrees of freedom
## AIC: 132.8
## 
## Number of Fisher Scoring iterations: 5
## 
##                        Estimate Std. Error z value        Pr(>|z|)
## (Intercept)              -2.923     0.4591  -6.368 0.0000000001918
## independentvariableYes    1.709     0.5282   3.235 0.0012163968156

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_responsible"
## [1] NA
##   variable yesno Freq
## 1       No    No   73
## 2      Yes    No   48
## 3       No   Yes   25
## 4      Yes   Yes   35

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 4.9, df = 1, p-value = 0.03
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value  Pr(>|z|)    
## (Intercept)              -1.072      0.232   -4.62 0.0000038 ***
## independentvariableYes    0.756      0.321    2.35     0.019 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 229.95  on 180  degrees of freedom
## Residual deviance: 224.32  on 179  degrees of freedom
## AIC: 228.3
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value    Pr(>|z|)
## (Intercept)             -1.0716     0.2317  -4.624 0.000003759
## independentvariableYes   0.7557     0.3211   2.354 0.018593119

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_responsible"
## [1] NA
##   variable yesno Freq
## 1       No    No   81
## 2      Yes    No   46
## 3       No   Yes   17
## 4      Yes   Yes   37

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 15, df = 1, p-value = 0.0001
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value     Pr(>|z|)    
## (Intercept)              -1.561      0.267   -5.85 0.0000000048 ***
## independentvariableYes    1.344      0.346    3.88       0.0001 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 220.62  on 180  degrees of freedom
## Residual deviance: 204.51  on 179  degrees of freedom
## AIC: 208.5
## 
## Number of Fisher Scoring iterations: 3
## 
##                        Estimate Std. Error z value       Pr(>|z|)
## (Intercept)              -1.561     0.2668  -5.853 0.000000004836
## independentvariableYes    1.344     0.3463   3.880 0.000104618445

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_responsible"
## [1] NA
##   variable yesno Freq
## 1       No    No   66
## 2      Yes    No   78
## 3       No   Yes   32
## 4      Yes   Yes    5

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 18, df = 1, p-value = 0.00002
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value Pr(>|z|)    
## (Intercept)              -0.724      0.215   -3.36  0.00078 ***
## independentvariableYes   -2.023      0.509   -3.97 0.000071 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 183.34  on 180  degrees of freedom
## Residual deviance: 161.60  on 179  degrees of freedom
## AIC: 165.6
## 
## Number of Fisher Scoring iterations: 5
## 
##                        Estimate Std. Error z value   Pr(>|z|)
## (Intercept)             -0.7239     0.2154  -3.361 0.00077758
## independentvariableYes  -2.0234     0.5091  -3.974 0.00007065

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_policy"
## [1] "Consider data privacy implications and take actions to protect data privacy"
##   variable yesno Freq
## 1       No    No   70
## 2      Yes    No   27
## 3       No   Yes   44
## 4      Yes   Yes   40

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 6.7, df = 1, p-value = 0.009
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value Pr(>|z|)   
## (Intercept)              -0.464      0.192   -2.41   0.0158 * 
## independentvariableYes    0.857      0.315    2.72   0.0064 **
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 249.98  on 180  degrees of freedom
## Residual deviance: 242.40  on 179  degrees of freedom
## AIC: 246.4
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value Pr(>|z|)
## (Intercept)             -0.4643     0.1924  -2.413 0.015805
## independentvariableYes   0.8573     0.3147   2.724 0.006447

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_policy"
## [1] "Ask about the data or model to understand potential limitations it has"
##   variable yesno Freq
## 1       No    No   78
## 2      Yes    No   19
## 3       No   Yes   36
## 4      Yes   Yes   48

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 26, df = 1, p-value = 0.0000004
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value   Pr(>|z|)    
## (Intercept)              -0.773      0.201   -3.84    0.00012 ***
## independentvariableYes    1.700      0.338    5.03 0.00000048 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 249.98  on 180  degrees of freedom
## Residual deviance: 222.10  on 179  degrees of freedom
## AIC: 226.1
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value     Pr(>|z|)
## (Intercept)             -0.7732     0.2015  -3.837 0.0001243646
## independentvariableYes   1.7000     0.3377   5.033 0.0000004818

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_policy"
## [1] "Conduct adversarial testing or red teaming"
##   variable yesno Freq
## 1       No    No   84
## 2      Yes    No   42
## 3       No   Yes   30
## 4      Yes   Yes   25

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 1.9, df = 1, p-value = 0.2
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value  Pr(>|z|)    
## (Intercept)              -1.030      0.213   -4.84 0.0000013 ***
## independentvariableYes    0.511      0.330    1.55      0.12    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 222.31  on 180  degrees of freedom
## Residual deviance: 219.92  on 179  degrees of freedom
## AIC: 223.9
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value    Pr(>|z|)
## (Intercept)             -1.0296     0.2127  -4.841 0.000001293
## independentvariableYes   0.5108     0.3302   1.547 0.121885215

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_policy"
## [1] "Use explainability methods"
##   variable yesno Freq
## 1       No    No   82
## 2      Yes    No   38
## 3       No   Yes   32
## 4      Yes   Yes   29

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 3.7, df = 1, p-value = 0.05
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value  Pr(>|z|)    
## (Intercept)              -0.941      0.208   -4.51 0.0000063 ***
## independentvariableYes    0.671      0.323    2.08     0.038 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 231.33  on 180  degrees of freedom
## Residual deviance: 227.01  on 179  degrees of freedom
## AIC: 231
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value    Pr(>|z|)
## (Intercept)             -0.9410     0.2084  -4.515 0.000006346
## independentvariableYes   0.6707     0.3229   2.077 0.037773949

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_policy"
## [1] "Build transparency approaches"
##   variable yesno Freq
## 1       No    No   70
## 2      Yes    No   31
## 3       No   Yes   44
## 4      Yes   Yes   36

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 3.3, df = 1, p-value = 0.07
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value Pr(>|z|)  
## (Intercept)              -0.464      0.192   -2.41    0.016 *
## independentvariableYes    0.614      0.312    1.97    0.049 *
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 248.48  on 180  degrees of freedom
## Residual deviance: 244.56  on 179  degrees of freedom
## AIC: 248.6
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value Pr(>|z|)
## (Intercept)             -0.4643     0.1924  -2.413  0.01581
## independentvariableYes   0.6138     0.3115   1.970  0.04879

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_policy"
## [1] "No actions taken"
##   variable yesno Freq
## 1       No    No   78
## 2      Yes    No   33
## 3       No   Yes   36
## 4      Yes   Yes   34

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 5.8, df = 1, p-value = 0.02
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value Pr(>|z|)    
## (Intercept)              -0.773      0.201   -3.84  0.00012 ***
## independentvariableYes    0.803      0.317    2.54  0.01123 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 241.55  on 180  degrees of freedom
## Residual deviance: 235.06  on 179  degrees of freedom
## AIC: 239.1
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value  Pr(>|z|)
## (Intercept)             -0.7732     0.2015  -3.837 0.0001244
## independentvariableYes   0.8030     0.3167   2.535 0.0112294

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_policy"
## [1] NA
##   variable yesno Freq
## 1       No    No  107
## 2      Yes    No   50
## 3       No   Yes    7
## 4      Yes   Yes   17

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 12, df = 1, p-value = 0.0005
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value        Pr(>|z|)    
## (Intercept)              -2.727      0.390   -6.99 0.0000000000028 ***
## independentvariableYes    1.648      0.481    3.43         0.00061 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 141.65  on 180  degrees of freedom
## Residual deviance: 128.52  on 179  degrees of freedom
## AIC: 132.5
## 
## Number of Fisher Scoring iterations: 5
## 
##                        Estimate Std. Error z value          Pr(>|z|)
## (Intercept)              -2.727     0.3901  -6.990 0.000000000002753
## independentvariableYes    1.648     0.4806   3.429 0.000605988717020

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_policy"
## [1] NA
##   variable yesno Freq
## 1       No    No   85
## 2      Yes    No   36
## 3       No   Yes   29
## 4      Yes   Yes   31

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 7.3, df = 1, p-value = 0.007
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value   Pr(>|z|)    
## (Intercept)              -1.075      0.215   -5.00 0.00000057 ***
## independentvariableYes    0.926      0.326    2.84     0.0045 ** 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 229.95  on 180  degrees of freedom
## Residual deviance: 221.81  on 179  degrees of freedom
## AIC: 225.8
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value    Pr(>|z|)
## (Intercept)             -1.0754     0.2151   -5.00 0.000000572
## independentvariableYes   0.9258     0.3260    2.84 0.004513427

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_policy"
## [1] NA
##   variable yesno Freq
## 1       No    No   87
## 2      Yes    No   40
## 3       No   Yes   27
## 4      Yes   Yes   27

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 4.8, df = 1, p-value = 0.03
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value   Pr(>|z|)    
## (Intercept)              -1.170      0.220   -5.31 0.00000011 ***
## independentvariableYes    0.777      0.333    2.34      0.019 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 220.62  on 180  degrees of freedom
## Residual deviance: 215.15  on 179  degrees of freedom
## AIC: 219.2
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value     Pr(>|z|)
## (Intercept)              -1.170     0.2203  -5.311 0.0000001088
## independentvariableYes    0.777     0.3325   2.337 0.0194495073

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_policy"
## [1] NA
##   variable yesno Freq
## 1       No    No   81
## 2      Yes    No   63
## 3       No   Yes   33
## 4      Yes   Yes    4

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 12, df = 1, p-value = 0.0004
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value Pr(>|z|)    
## (Intercept)              -0.898      0.207   -4.35 0.000014 ***
## independentvariableYes   -1.859      0.555   -3.35  0.00082 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 183.34  on 180  degrees of freedom
## Residual deviance: 167.49  on 179  degrees of freedom
## AIC: 171.5
## 
## Number of Fisher Scoring iterations: 5
## 
##                        Estimate Std. Error z value   Pr(>|z|)
## (Intercept)             -0.8979     0.2065  -4.348 0.00001373
## independentvariableYes  -1.8589     0.5554  -3.347 0.00081780

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_incentives"
## [1] "Consider data privacy implications and take actions to protect data privacy"
##   variable yesno Freq
## 1       No    No   82
## 2      Yes    No   15
## 3       No   Yes   60
## 4      Yes   Yes   24

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 3.8, df = 1, p-value = 0.05
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value Pr(>|z|)  
## (Intercept)              -0.312      0.170   -1.84    0.066 .
## independentvariableYes    0.782      0.370    2.11    0.035 *
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 249.98  on 180  degrees of freedom
## Residual deviance: 245.40  on 179  degrees of freedom
## AIC: 249.4
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value Pr(>|z|)
## (Intercept)             -0.3124     0.1699  -1.839  0.06596
## independentvariableYes   0.7824     0.3704   2.112  0.03466

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_incentives"
## [1] "Ask about the data or model to understand potential limitations it has"
##   variable yesno Freq
## 1       No    No   84
## 2      Yes    No   13
## 3       No   Yes   58
## 4      Yes   Yes   26

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 7.2, df = 1, p-value = 0.007
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value Pr(>|z|)   
## (Intercept)              -0.370      0.171   -2.17   0.0300 * 
## independentvariableYes    1.064      0.380    2.80   0.0052 **
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 249.98  on 180  degrees of freedom
## Residual deviance: 241.71  on 179  degrees of freedom
## AIC: 245.7
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value Pr(>|z|)
## (Intercept)             -0.3704     0.1707  -2.169  0.03005
## independentvariableYes   1.0635     0.3802   2.797  0.00515

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_incentives"
## [1] "Conduct adversarial testing or red teaming"
##   variable yesno Freq
## 1       No    No  109
## 2      Yes    No   17
## 3       No   Yes   33
## 4      Yes   Yes   22

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 14, df = 1, p-value = 0.0001
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value     Pr(>|z|)    
## (Intercept)              -1.195      0.199   -6.01 0.0000000018 ***
## independentvariableYes    1.453      0.379    3.83      0.00013 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 222.31  on 180  degrees of freedom
## Residual deviance: 207.39  on 179  degrees of freedom
## AIC: 211.4
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value       Pr(>|z|)
## (Intercept)              -1.195     0.1987  -6.014 0.000000001814
## independentvariableYes    1.453     0.3792   3.831 0.000127428575

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_incentives"
## [1] "Use explainability methods"
##   variable yesno Freq
## 1       No    No   98
## 2      Yes    No   22
## 3       No   Yes   44
## 4      Yes   Yes   17

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 1.6, df = 1, p-value = 0.2
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value Pr(>|z|)    
## (Intercept)              -0.801      0.181   -4.41  0.00001 ***
## independentvariableYes    0.543      0.370    1.47     0.14    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 231.33  on 180  degrees of freedom
## Residual deviance: 229.22  on 179  degrees of freedom
## AIC: 233.2
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value   Pr(>|z|)
## (Intercept)             -0.8008     0.1815  -4.413 0.00001021
## independentvariableYes   0.5429     0.3704   1.466 0.14270998

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_incentives"
## [1] "Build transparency approaches"
##   variable yesno Freq
## 1       No    No   82
## 2      Yes    No   19
## 3       No   Yes   60
## 4      Yes   Yes   20

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 0.68, df = 1, p-value = 0.4
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value Pr(>|z|)  
## (Intercept)              -0.312      0.170   -1.84    0.066 .
## independentvariableYes    0.364      0.363    1.00    0.316  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 248.48  on 180  degrees of freedom
## Residual deviance: 247.47  on 179  degrees of freedom
## AIC: 251.5
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value Pr(>|z|)
## (Intercept)             -0.3124     0.1699  -1.839  0.06596
## independentvariableYes   0.3637     0.3626   1.003  0.31591

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_incentives"
## [1] "No actions taken"
##   variable yesno Freq
## 1       No    No   93
## 2      Yes    No   18
## 3       No   Yes   49
## 4      Yes   Yes   21

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 4, df = 1, p-value = 0.04
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value Pr(>|z|)    
## (Intercept)              -0.641      0.177   -3.63  0.00028 ***
## independentvariableYes    0.795      0.367    2.17  0.03009 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 241.55  on 180  degrees of freedom
## Residual deviance: 236.83  on 179  degrees of freedom
## AIC: 240.8
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value  Pr(>|z|)
## (Intercept)             -0.6408     0.1765  -3.630 0.0002834
## independentvariableYes   0.7949     0.3665   2.169 0.0300926

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_incentives"
## [1] NA
##   variable yesno Freq
## 1       No    No  130
## 2      Yes    No   27
## 3       No   Yes   12
## 4      Yes   Yes   12

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 11, df = 1, p-value = 0.0007
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value           Pr(>|z|)    
## (Intercept)              -2.383      0.302   -7.90 0.0000000000000029 ***
## independentvariableYes    1.572      0.460    3.42            0.00063 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 141.65  on 180  degrees of freedom
## Residual deviance: 130.40  on 179  degrees of freedom
## AIC: 134.4
## 
## Number of Fisher Scoring iterations: 5
## 
##                        Estimate Std. Error z value             Pr(>|z|)
## (Intercept)              -2.383     0.3017  -7.897 0.000000000000002852
## independentvariableYes    1.572     0.4598   3.418 0.000629945814988855

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_incentives"
## [1] NA
##   variable yesno Freq
## 1       No    No   98
## 2      Yes    No   23
## 3       No   Yes   44
## 4      Yes   Yes   16

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 0.98, df = 1, p-value = 0.3
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value Pr(>|z|)    
## (Intercept)              -0.801      0.181   -4.41  0.00001 ***
## independentvariableYes    0.438      0.373    1.17     0.24    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 229.95  on 180  degrees of freedom
## Residual deviance: 228.59  on 179  degrees of freedom
## AIC: 232.6
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value   Pr(>|z|)
## (Intercept)             -0.8008     0.1815  -4.413 0.00001021
## independentvariableYes   0.4379     0.3727   1.175 0.24005583

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_incentives"
## [1] NA
##   variable yesno Freq
## 1       No    No  106
## 2      Yes    No   21
## 3       No   Yes   36
## 4      Yes   Yes   18

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 5.4, df = 1, p-value = 0.02
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value    Pr(>|z|)    
## (Intercept)              -1.080      0.193   -5.60 0.000000022 ***
## independentvariableYes    0.926      0.375    2.47       0.013 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 220.62  on 180  degrees of freedom
## Residual deviance: 214.63  on 179  degrees of freedom
## AIC: 218.6
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value      Pr(>|z|)
## (Intercept)             -1.0799     0.1929  -5.598 0.00000002165
## independentvariableYes   0.9258     0.3747   2.471 0.01348049011

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_incentives"
## [1] NA
##   variable yesno Freq
## 1       No    No  108
## 2      Yes    No   36
## 3       No   Yes   34
## 4      Yes   Yes    3

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 4, df = 1, p-value = 0.04
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value     Pr(>|z|)    
## (Intercept)              -1.156      0.197   -5.88 0.0000000042 ***
## independentvariableYes   -1.329      0.632   -2.10        0.036 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 183.34  on 180  degrees of freedom
## Residual deviance: 177.47  on 179  degrees of freedom
## AIC: 181.5
## 
## Number of Fisher Scoring iterations: 5
## 
##                        Estimate Std. Error z value      Pr(>|z|)
## (Intercept)              -1.156     0.1966  -5.877 0.00000000417
## independentvariableYes   -1.329     0.6323  -2.102 0.03554258538

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_neither"
## [1] "Consider data privacy implications and take actions to protect data privacy"
##   variable yesno Freq
## 1       No    No   68
## 2      Yes    No   29
## 3       No   Yes   76
## 4      Yes   Yes    8

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 10, df = 1, p-value = 0.001
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value Pr(>|z|)   
## (Intercept)               0.111      0.167    0.67   0.5052   
## independentvariableYes   -1.399      0.433   -3.23   0.0012 **
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 249.98  on 180  degrees of freedom
## Residual deviance: 237.82  on 179  degrees of freedom
## AIC: 241.8
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value Pr(>|z|)
## (Intercept)              0.1112     0.1669  0.6663 0.505205
## independentvariableYes  -1.3991     0.4328 -3.2324 0.001228

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_neither"
## [1] "Ask about the data or model to understand potential limitations it has"
##   variable yesno Freq
## 1       No    No   62
## 2      Yes    No   35
## 3       No   Yes   82
## 4      Yes   Yes    2

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 29, df = 1, p-value = 0.00000006
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value Pr(>|z|)    
## (Intercept)               0.280      0.168    1.66    0.097 .  
## independentvariableYes   -3.142      0.746   -4.21 0.000026 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 249.98  on 180  degrees of freedom
## Residual deviance: 212.40  on 179  degrees of freedom
## AIC: 216.4
## 
## Number of Fisher Scoring iterations: 5
## 
##                        Estimate Std. Error z value   Pr(>|z|)
## (Intercept)              0.2796     0.1683   1.661 0.09666311
## independentvariableYes  -3.1418     0.7462  -4.210 0.00002552

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_neither"
## [1] "Conduct adversarial testing or red teaming"
##   variable yesno Freq
## 1       No    No   96
## 2      Yes    No   30
## 3       No   Yes   48
## 4      Yes   Yes    7

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 2.3, df = 1, p-value = 0.1
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value Pr(>|z|)    
## (Intercept)              -0.693      0.177   -3.92 0.000088 ***
## independentvariableYes   -0.762      0.455   -1.67    0.094 .  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 222.31  on 180  degrees of freedom
## Residual deviance: 219.21  on 179  degrees of freedom
## AIC: 223.2
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value   Pr(>|z|)
## (Intercept)             -0.6931     0.1768  -3.921 0.00008817
## independentvariableYes  -0.7621     0.4555  -1.673 0.09425743

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_neither"
## [1] "Use explainability methods"
##   variable yesno Freq
## 1       No    No   89
## 2      Yes    No   31
## 3       No   Yes   55
## 4      Yes   Yes    6

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 5.4, df = 1, p-value = 0.02
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value Pr(>|z|)   
## (Intercept)              -0.481      0.172   -2.81    0.005 **
## independentvariableYes   -1.161      0.478   -2.43    0.015 * 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 231.33  on 180  degrees of freedom
## Residual deviance: 224.32  on 179  degrees of freedom
## AIC: 228.3
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value Pr(>|z|)
## (Intercept)             -0.4813     0.1715  -2.806 0.005013
## independentvariableYes  -1.1609     0.4779  -2.429 0.015121

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_neither"
## [1] "Build transparency approaches"
##   variable yesno Freq
## 1       No    No   82
## 2      Yes    No   19
## 3       No   Yes   62
## 4      Yes   Yes   18

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 0.18, df = 1, p-value = 0.7
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value Pr(>|z|)  
## (Intercept)              -0.280      0.168   -1.66    0.097 .
## independentvariableYes    0.226      0.369    0.61    0.542  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 248.48  on 180  degrees of freedom
## Residual deviance: 248.11  on 179  degrees of freedom
## AIC: 252.1
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value Pr(>|z|)
## (Intercept)             -0.2796     0.1683 -1.6613  0.09666
## independentvariableYes   0.2255     0.3695  0.6104  0.54161

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_neither"
## [1] "No actions taken"
##   variable yesno Freq
## 1       No    No   84
## 2      Yes    No   27
## 3       No   Yes   60
## 4      Yes   Yes   10

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 2.1, df = 1, p-value = 0.1
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value Pr(>|z|)  
## (Intercept)              -0.336      0.169   -1.99    0.047 *
## independentvariableYes   -0.657      0.407   -1.61    0.107  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 241.55  on 180  degrees of freedom
## Residual deviance: 238.79  on 179  degrees of freedom
## AIC: 242.8
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value Pr(>|z|)
## (Intercept)             -0.3365     0.1690  -1.991  0.04653
## independentvariableYes  -0.6568     0.4069  -1.614  0.10655

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_neither"
## [1] NA
##   variable yesno Freq
## 1       No    No  122
## 2      Yes    No   35
## 3       No   Yes   22
## 4      Yes   Yes    2

## Warning in chisq.test(table(cbind(as.data.frame(orgsresponsiblegenai[iv]), : Chi-squared approximation may be incorrect

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 1.7, df = 1, p-value = 0.2
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value         Pr(>|z|)    
## (Intercept)              -1.713      0.232   -7.40 0.00000000000014 ***
## independentvariableYes   -1.149      0.763   -1.51             0.13    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 141.65  on 180  degrees of freedom
## Residual deviance: 138.68  on 179  degrees of freedom
## AIC: 142.7
## 
## Number of Fisher Scoring iterations: 5
## 
##                        Estimate Std. Error z value          Pr(>|z|)
## (Intercept)              -1.713     0.2316  -7.395 0.000000000000141
## independentvariableYes   -1.149     0.7630  -1.506 0.132027157691305

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_neither"
## [1] NA
##   variable yesno Freq
## 1       No    No   89
## 2      Yes    No   32
## 3       No   Yes   55
## 4      Yes   Yes    5

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 7, df = 1, p-value = 0.008
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value Pr(>|z|)   
## (Intercept)              -0.481      0.172   -2.81   0.0050 **
## independentvariableYes   -1.375      0.511   -2.69   0.0071 **
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 229.95  on 180  degrees of freedom
## Residual deviance: 220.83  on 179  degrees of freedom
## AIC: 224.8
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value Pr(>|z|)
## (Intercept)             -0.4813     0.1715  -2.806 0.005013
## independentvariableYes  -1.3750     0.5106  -2.693 0.007078

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_neither"
## [1] NA
##   variable yesno Freq
## 1       No    No   93
## 2      Yes    No   34
## 3       No   Yes   51
## 4      Yes   Yes    3

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 9.2, df = 1, p-value = 0.002
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value Pr(>|z|)    
## (Intercept)              -0.601      0.174   -3.45  0.00056 ***
## independentvariableYes   -1.827      0.627   -2.91  0.00357 ** 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 220.62  on 180  degrees of freedom
## Residual deviance: 208.02  on 179  degrees of freedom
## AIC: 212
## 
## Number of Fisher Scoring iterations: 5
## 
##                        Estimate Std. Error z value  Pr(>|z|)
## (Intercept)             -0.6008     0.1742  -3.448 0.0005649
## independentvariableYes  -1.8270     0.6270  -2.914 0.0035692

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`

## [1] "orghave_neither"
## [1] NA
##   variable yesno Freq
## 1       No    No  124
## 2      Yes    No   20
## 3       No   Yes   20
## 4      Yes   Yes   17

## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  table(cbind(as.data.frame(orgsresponsiblegenai[iv]), as.data.frame(orgsresponsiblegenai[column])))
## X-squared = 17, df = 1, p-value = 0.00004
## 
## 
## Call:
## glm(formula = yvariable ~ independentvariable, family = "binomial", 
##     data = orgsresponsiblegenai)
## 
## Coefficients:
##                        Estimate Std. Error z value          Pr(>|z|)    
## (Intercept)              -1.825      0.241   -7.57 0.000000000000037 ***
## independentvariableYes    1.662      0.409    4.07 0.000047331845012 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 183.34  on 180  degrees of freedom
## Residual deviance: 167.10  on 179  degrees of freedom
## AIC: 171.1
## 
## Number of Fisher Scoring iterations: 4
## 
##                        Estimate Std. Error z value           Pr(>|z|)
## (Intercept)              -1.825     0.2410  -7.572 0.0000000000000368
## independentvariableYes    1.662     0.4085   4.068 0.0000473318450125

## Waiting for profiling to be done...
## Joining with `by = join_by(rowname)`