7283_HW3

Survey Data Source: National Household Education Surveys (NHES) Program 2019: Parent and Family Involvement in Education (PFI)

Research question

How do the factors of parent volunteerism, developmental delay, parent’s highest education level, and race/ethnicity affect whether a child enjoys school?

Predictor variables

Predictor 1: adult_volunteer; Item 60B: FSVOL “… has any adult in this child’s household … served as a volunteer in this child’s classroom or elsewhere in the school?”

Predictor 2: dev_delay; Item 76K: HDDELAYX “Has a health or education professional told you that this child has … a developmental delay?”

Predictor 3: parent_educ; PARGRADEX “Parent/guardian highest education”

Predictor 4: race_eth; RACEETH “Race and ethnicity of child”

Results

Present results from a model with sample weights and design effects, if your data allow for this. Present the results in tabular form, with Parameter estimates, odds ratios (if using the logit model) and confidence intervals for the odds ratios.

library(srvyr)

options(survey.lonely.psu = "adjust")

pfi19design <- svydesign(ids = ~PPSU, strata= ~PSTRATUM, weights = ~FPWT, data = pfi19, nest = TRUE)
pfi19design

## Stratified Independent Sampling design (with replacement)
## svydesign(ids = ~PPSU, strata = ~PSTRATUM, weights = ~FPWT, data = pfi19, 
##     nest = TRUE)

fit.logit <- svyglm(enjoy_school ~ adult_volunteer + dev_delay + parent_educ + race_eth, 
                    design = pfi19design, 
                    family = binomial)

## Warning in eval(family$initialize): non-integer #successes in a binomial glm!

summary(fit.logit)

## 
## Call:
## svyglm(formula = enjoy_school ~ adult_volunteer + dev_delay + 
##     parent_educ + race_eth, design = pfi19design, family = binomial)
## 
## Survey design:
## svydesign(ids = ~PPSU, strata = ~PSTRATUM, weights = ~FPWT, data = pfi19, 
##     nest = TRUE)
## 
## Coefficients:
##                          Estimate Std. Error t value Pr(>|t|)    
## (Intercept)               2.06369    0.20550  10.042  < 2e-16 ***
## adult_volunteerYes        0.69128    0.07698   8.980  < 2e-16 ***
## dev_delayYes             -0.91275    0.15752  -5.795 6.98e-09 ***
## parent_educ1HS Grad      -0.31477    0.21816  -1.443  0.14909    
## parent_educ2Some College -0.30100    0.20590  -1.462  0.14379    
## parent_educ3College Grad -0.25541    0.21493  -1.188  0.23472    
## parent_educ4Grad School  -0.04690    0.22242  -0.211  0.83301    
## race_ethHispanic          0.32252    0.10356   3.114  0.00185 ** 
## race_ethNH Asian          0.44188    0.27674   1.597  0.11034    
## race_ethNH Black          0.36178    0.13751   2.631  0.00852 ** 
## race_ethOther            -0.17170    0.13944  -1.231  0.21819    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 0.9963352)
## 
## Number of Fisher Scoring iterations: 5

library(gtsummary)
fit.logit %>%
  tbl_regression(exponentiate=TRUE)

Characteristic	OR1	95% CI1	p-value
adult_volunteer
No	—	—
Yes	2.00	1.72, 2.32	<0.001
dev_delay
No	—	—
Yes	0.40	0.29, 0.55	<0.001
parent_educ
0Less than HS	—	—
1HS Grad	0.73	0.48, 1.12	0.15
2Some College	0.74	0.49, 1.11	0.14
3College Grad	0.77	0.51, 1.18	0.2
4Grad School	0.95	0.62, 1.48	0.8
race_eth
NH White	—	—
Hispanic	1.38	1.13, 1.69	0.002
NH Asian	1.56	0.90, 2.68	0.11
NH Black	1.44	1.10, 1.88	0.009
Other	0.84	0.64, 1.11	0.2
1 OR = Odds Ratio, CI = Confidence Interval

library(sjPlot)
plot_model(fit.logit,
           axis.lim = c(.1, 10), 
           title = "Odds Ratios for Children Enjoying School")

Interesting cases

Generate predicted probabilities for some “interesting” cases from your analysis, to highlight the effects from the model and your stated research question.

library(emmeans)
rg <- ref_grid(fit.logit)

marg_logit <- emmeans(object = rg,
              specs = c("adult_volunteer",  "dev_delay", "race_eth", "parent_educ"),
              type = "response")

knitr::kable(marg_logit, digits = 4)

adult_volunteer	dev_delay	race_eth	parent_educ	prob	SE	df	asymp.LCL	asymp.UCL
No	No	NH White	0Less than HS	0.8873	0.0205	Inf	0.8404	0.9218
Yes	No	NH White	0Less than HS	0.9402	0.0118	Inf	0.9125	0.9595
No	Yes	NH White	0Less than HS	0.7597	0.0469	Inf	0.6565	0.8394
Yes	Yes	NH White	0Less than HS	0.8632	0.0312	Inf	0.7898	0.9138
No	No	Hispanic	0Less than HS	0.9158	0.0144	Inf	0.8830	0.9400
Yes	No	Hispanic	0Less than HS	0.9560	0.0082	Inf	0.9367	0.9696
No	Yes	Hispanic	0Less than HS	0.8136	0.0383	Inf	0.7268	0.8774
Yes	Yes	Hispanic	0Less than HS	0.8970	0.0244	Inf	0.8385	0.9360
No	No	NH Asian	0Less than HS	0.9245	0.0297	Inf	0.8419	0.9657
Yes	No	NH Asian	0Less than HS	0.9607	0.0160	Inf	0.9140	0.9825
No	Yes	NH Asian	0Less than HS	0.8310	0.0629	Inf	0.6714	0.9221
Yes	Yes	NH Asian	0Less than HS	0.9076	0.0378	Inf	0.8023	0.9596
No	No	NH Black	0Less than HS	0.9187	0.0167	Inf	0.8793	0.9461
Yes	No	NH Black	0Less than HS	0.9576	0.0092	Inf	0.9353	0.9724
No	Yes	NH Black	0Less than HS	0.8195	0.0396	Inf	0.7287	0.8847
Yes	Yes	NH Black	0Less than HS	0.9006	0.0246	Inf	0.8410	0.9395
No	No	Other	0Less than HS	0.8690	0.0262	Inf	0.8086	0.9124
Yes	No	Other	0Less than HS	0.9298	0.0153	Inf	0.8931	0.9545
No	Yes	Other	0Less than HS	0.7270	0.0553	Inf	0.6067	0.8213
Yes	Yes	Other	0Less than HS	0.8416	0.0382	Inf	0.7519	0.9031
No	No	NH White	1HS Grad	0.8518	0.0139	Inf	0.8224	0.8771
Yes	No	NH White	1HS Grad	0.9198	0.0089	Inf	0.9007	0.9356
No	Yes	NH White	1HS Grad	0.6977	0.0373	Inf	0.6199	0.7655
Yes	Yes	NH White	1HS Grad	0.8216	0.0278	Inf	0.7606	0.8698
No	No	Hispanic	1HS Grad	0.8881	0.0122	Inf	0.8618	0.9099
Yes	No	Hispanic	1HS Grad	0.9406	0.0078	Inf	0.9234	0.9542
No	Yes	Hispanic	1HS Grad	0.7611	0.0362	Inf	0.6831	0.8248
Yes	Yes	Hispanic	1HS Grad	0.8641	0.0253	Inf	0.8066	0.9065
No	No	NH Asian	1HS Grad	0.8994	0.0266	Inf	0.8340	0.9409
Yes	No	NH Asian	1HS Grad	0.9470	0.0148	Inf	0.9091	0.9696
No	Yes	NH Asian	1HS Grad	0.7821	0.0544	Inf	0.6575	0.8703
Yes	Yes	NH Asian	1HS Grad	0.8775	0.0348	Inf	0.7915	0.9312
No	No	NH Black	1HS Grad	0.8919	0.0151	Inf	0.8585	0.9182
Yes	No	NH Black	1HS Grad	0.9428	0.0088	Inf	0.9229	0.9578
No	Yes	NH Black	1HS Grad	0.7682	0.0362	Inf	0.6898	0.8316
Yes	Yes	NH Black	1HS Grad	0.8687	0.0244	Inf	0.8131	0.9096
No	No	Other	1HS Grad	0.8288	0.0233	Inf	0.7782	0.8698
Yes	No	Other	1HS Grad	0.9062	0.0147	Inf	0.8732	0.9313
No	Yes	Other	1HS Grad	0.6603	0.0487	Inf	0.5594	0.7484
Yes	Yes	Other	1HS Grad	0.7951	0.0373	Inf	0.7125	0.8586
No	No	NH White	2Some College	0.8535	0.0096	Inf	0.8338	0.8713
Yes	No	NH White	2Some College	0.9209	0.0067	Inf	0.9067	0.9330
No	Yes	NH White	2Some College	0.7006	0.0328	Inf	0.6325	0.7607
Yes	Yes	NH White	2Some College	0.8236	0.0249	Inf	0.7694	0.8673
No	No	Hispanic	2Some College	0.8895	0.0099	Inf	0.8685	0.9074
Yes	No	Hispanic	2Some College	0.9414	0.0067	Inf	0.9268	0.9532
No	Yes	Hispanic	2Some College	0.7636	0.0334	Inf	0.6921	0.8227
Yes	Yes	Hispanic	2Some College	0.8657	0.0236	Inf	0.8125	0.9056
No	No	NH Asian	2Some College	0.9007	0.0255	Inf	0.8384	0.9406
Yes	No	NH Asian	2Some College	0.9476	0.0142	Inf	0.9117	0.9695
No	Yes	NH Asian	2Some College	0.7845	0.0524	Inf	0.6647	0.8698
Yes	Yes	NH Asian	2Some College	0.8790	0.0335	Inf	0.7965	0.9309
No	No	NH Black	2Some College	0.8933	0.0132	Inf	0.8644	0.9166
Yes	No	NH Black	2Some College	0.9435	0.0078	Inf	0.9261	0.9570
No	Yes	NH Black	2Some College	0.7706	0.0333	Inf	0.6989	0.8294
Yes	Yes	NH Black	2Some College	0.8702	0.0226	Inf	0.8191	0.9085
No	No	Other	2Some College	0.8308	0.0197	Inf	0.7887	0.8659
Yes	No	Other	2Some College	0.9074	0.0127	Inf	0.8794	0.9294
No	Yes	Other	2Some College	0.6633	0.0442	Inf	0.5721	0.7439
Yes	Yes	Other	2Some College	0.7973	0.0341	Inf	0.7222	0.8561
No	No	NH White	3College Grad	0.8592	0.0094	Inf	0.8397	0.8766
Yes	No	NH White	3College Grad	0.9241	0.0055	Inf	0.9125	0.9343
No	Yes	NH White	3College Grad	0.7100	0.0335	Inf	0.6402	0.7711
Yes	Yes	NH White	3College Grad	0.8302	0.0240	Inf	0.7778	0.8722
No	No	Hispanic	3College Grad	0.8939	0.0107	Inf	0.8710	0.9131
Yes	No	Hispanic	3College Grad	0.9439	0.0065	Inf	0.9297	0.9553
No	Yes	Hispanic	3College Grad	0.7717	0.0346	Inf	0.6970	0.8325
Yes	Yes	Hispanic	3College Grad	0.8709	0.0233	Inf	0.8180	0.9102
No	No	NH Asian	3College Grad	0.9047	0.0237	Inf	0.8470	0.9421
Yes	No	NH Asian	3College Grad	0.9499	0.0130	Inf	0.9174	0.9700
No	Yes	NH Asian	3College Grad	0.7921	0.0501	Inf	0.6773	0.8736
Yes	Yes	NH Asian	3College Grad	0.8838	0.0314	Inf	0.8068	0.9326
No	No	NH Black	3College Grad	0.8975	0.0140	Inf	0.8666	0.9220
Yes	No	NH Black	3College Grad	0.9459	0.0078	Inf	0.9284	0.9593
No	Yes	NH Black	3College Grad	0.7786	0.0351	Inf	0.7024	0.8397
Yes	Yes	NH Black	3College Grad	0.8753	0.0228	Inf	0.8234	0.9136
No	No	Other	3College Grad	0.8371	0.0198	Inf	0.7945	0.8722
Yes	No	Other	3College Grad	0.9112	0.0119	Inf	0.8848	0.9319
No	Yes	Other	3College Grad	0.6734	0.0453	Inf	0.5793	0.7554
Yes	Yes	Other	3College Grad	0.8046	0.0335	Inf	0.7305	0.8621
No	No	NH White	4Grad School	0.8825	0.0086	Inf	0.8645	0.8985
Yes	No	NH White	4Grad School	0.9375	0.0050	Inf	0.9269	0.9467
No	Yes	NH White	4Grad School	0.7510	0.0316	Inf	0.6842	0.8077
Yes	Yes	NH White	4Grad School	0.8576	0.0216	Inf	0.8098	0.8949
No	No	Hispanic	4Grad School	0.9121	0.0098	Inf	0.8909	0.9295
Yes	No	Hispanic	4Grad School	0.9539	0.0058	Inf	0.9412	0.9640
No	Yes	Hispanic	4Grad School	0.8064	0.0319	Inf	0.7361	0.8615
Yes	Yes	Hispanic	4Grad School	0.8926	0.0207	Inf	0.8449	0.9269
No	No	NH Asian	4Grad School	0.9212	0.0194	Inf	0.8738	0.9518
Yes	No	NH Asian	4Grad School	0.9589	0.0104	Inf	0.9328	0.9751
No	Yes	NH Asian	4Grad School	0.8243	0.0433	Inf	0.7231	0.8940
Yes	Yes	NH Asian	4Grad School	0.9035	0.0262	Inf	0.8385	0.9441
No	No	NH Black	4Grad School	0.9152	0.0118	Inf	0.8890	0.9356
Yes	No	NH Black	4Grad School	0.9556	0.0065	Inf	0.9411	0.9667
No	Yes	NH Black	4Grad School	0.8124	0.0313	Inf	0.7433	0.8663
Yes	Yes	NH Black	4Grad School	0.8963	0.0196	Inf	0.8510	0.9290
No	No	Other	4Grad School	0.8635	0.0185	Inf	0.8231	0.8959
Yes	No	Other	4Grad School	0.9267	0.0109	Inf	0.9023	0.9453
No	Yes	Other	4Grad School	0.7175	0.0440	Inf	0.6240	0.7954
Yes	Yes	Other	4Grad School	0.8353	0.0309	Inf	0.7656	0.8873

comps <- as.data.frame(marg_logit)

comps[comps$adult_volunteer == "Yes" & comps$dev_delay == "Yes", ]

comps[comps$dev_delay == "No" , ]

Interestingly, there were no significant effects based on parent/guardian education gradient or child race/ethnicity. Across those categories, parent volunteerism increased the likelihood of a child enjoying school and diagnosis of a developmental delay decreased the likelihood of a child enjoying school.