Differences in response rate by demographic and other characteristics

In this document, we examine differences in response rate for STEM-IE by:

We find that there are URM and gender differences in response rate, which appear to be substantively important, but that there are not statistically significant differences in URM status or gender.

Race

There are not statistically significant differences by race. Other includes white (n = 13), multiracial (n = 5), and Asian (n = 14).

## # A tibble: 4 x 2
##       race mean_response_rate
##     <fctr>              <dbl>
## 1    Black          0.6025861
## 2 Hispanic          0.6142559
## 3    Other          0.6918350
## 4     <NA>          0.7727273
##              Df Sum Sq Mean Sq F value Pr(>F)
## race          2  0.178 0.08911   1.377  0.255
## Residuals   184 11.910 0.06473               
## 2 observations deleted due to missingness

URM

Students who are in a URM group have a response rate of 60.6% compared to a response rate of 74.1% for non-URM students (t = 2.622, p = .013, d = .56).

## # A tibble: 2 x 2
##     urm mean_response_rate
##   <dbl>              <dbl>
## 1     0          0.7409596
## 2     1          0.6060130
## [1] "mean in group 0  is  0.741"
## [1] "mean in group 1  is  0.606"
## [1] "Test statistic is  2.622"
## [1] "P-value is  0.013"
## [1] "Effect size is  0.56"
## 
##  Welch Two Sample t-test
## 
## data:  ddd$response_rate by ddd$urm
## t = 2.6217, df = 32.92, p-value = 0.01314
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  0.03021329 0.23967985
## sample estimates:
## mean in group 0 mean in group 1 
##       0.7409596       0.6060130

Gender

Female’s response rate is 66.2% compared to male’s, which is 58.4%; this is a significant difference (t = 2.093, p = .038, d = .31)

## # A tibble: 3 x 2
##   gender mean_response_rate
##    <chr>              <dbl>
## 1      F          0.6615822
## 2      M          0.5837968
## 3   <NA>          0.6363636
## [1] "mean in group F  is  0.662"
## [1] "mean in group M  is  0.584"
## [1] "Test statistic is  2.093"
## [1] "P-value is  0.038"
## [1] "Effect size is  0.31"
## 
##  Welch Two Sample t-test
## 
## data:  ddd$response_rate by ddd$gender
## t = 2.0935, df = 184.61, p-value = 0.03767
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  0.004480866 0.151090081
## sample estimates:
## mean in group F mean in group M 
##       0.6615822       0.5837968

Pre-interest

Overall pre-interest and response rate are not correlated.

## 
##  Pearson's product-moment correlation
## 
## data:  ddd$overall_pre_interest and ddd$response_rate
## t = -0.032217, df = 167, p-value = 0.9743
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
##  -0.1533953  0.1485229
## sample estimates:
##          cor 
## -0.002493016