Fan et al. 2017Online resource 2,page 1
Online Module for Carrier Screening in Ashkenazi Jewish Individuals Compared With In-Person Genetics Education: A Randomized Controlled Trial.
Journal of Genetic Counseling
FAN, Chia Wei1*; CASTONGUAY, Lysanne1*; RUMMELL, Sonja1*; LÉVESQUE, Sébastien2; MITCHELL, John J.1,3,4; SILLON, Guillaume1,3†
1. Department of Human Genetics, Faculty of Medicine, McGill University, Montreal, Quebec, Canada.
2. Department of Pediatrics, Division of Medical Genetics, Faculty of Medicine and Health Sciences, Université de Sherbrooke, Sherbrooke, Quebec, Canada
3. Department of Medical Genetics, McGill University Health Center, Montreal, Quebec, Canada
4. Department of Endocrinology and Metabolism, McGill University Health Center, Montreal, Quebec, Canada
* These authors have contributed equally.
† Correspondence to:
Guillaume Sillon
Department of Medical Genetics, McGill University Health Centre
1001 boul. Décarie, Room A04.3140
Montreal, Quebec, Canada H4A 3J1
Tel: +1-514-412-4427. Fax: +1-514-412-4296.
Email:
Fan et al. 2017Online resource 2,page 1
Online resource 2: ANCOVA Linear Model Summary Data
CoefficientsVariable / Estimate (95% CI) / SE / t value / p value / Residual SE (df = 51) / Multiple R2 / Adjusted R2 / F statistic (df = 2 and 51) / p value
Knowledge / 1.0 / 0.29 / 0.27 / 11 / 0.00014
Intercept / 7.8 (7.0–8.6) / 0.40 / 20 / < 2.0e-16
Pre-education / 0.24 (0.13–0.35) / 0.053 / 4.5 / 3.5e-5
Treatment / -0.18 (-0.74–0.38) / 0.28 / -0.65 / 0.52
Risk perception / 0.48 / 0.39 / 0.36 / 16 / 3.9e-6
Intercept / 0.68 (0.24–1.1) / 0.22 / 3.1 / 0.0033
Pre-education / 0.63 (0.40–0.85) / 0.11 / 5.7 / 6.8e-7
Treatment / 0.081 (-0.18–0.35) / 0.13 / 0.62 / 0.54
Anxiety / 2.3 / 0.46 / 0.44 / 22 / 1.2e-7
Intercept / 2.8 (0.89–4.8) / 0.97 / 2.9 / 0.0052
Pre-education / 0.64 (0.45–0.83) / 0.096 / 6.6 / 2.1e-8
Treatment / 0.42 (-0.82–1.6) / 0.61 / 0.68 / 0.50
For all three outcome measures, the data were produced using the linear model function in R (lm(post ~ pre + treatment)). This solves for:
y = bxx + btt + a
where
y = post-education scoresx = covariate = pre-education scorest = treatment (t = 1 if web-based; t = 0 if in-person)
a = interceptbx = coefficient of the covariate = effect size of pre-education scores
and bt = coefficient of the treatment = effect size of web-based treatment
Fan et al. 2017Online resource 2,page 1
ANCOVA Raw Results: R Workspace Input and Output Without Formatting
R version 3.1.2 (2014-10-31) -- "Pumpkin Helmet"
Copyright (C) 2014 The R Foundation for Statistical Computing
Platform: x86_64-w64-mingw32/x64 (64-bit)
R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
Type 'license()' or 'licence()' for distribution details.
Natural language support but running in an English locale
R is a collaborative project with many contributors.
Type 'contributors()' for more information and
'citation()' on how to cite R or R packages in publications.
Type 'demo()' for some demos, 'help()' for on-line help, or
'help.start()' for an HTML browser interface to help.
Type 'q()' to quit R.
[Previously saved workspace restored]
> #### example code for ANCOVA ###
> #### ANCOVA (March 28, 2015)
> rm(list=ls.str())
> ## read data ##
> mydata <- matrix(
+ c(1,0,5,10,10,12,2,2,
+ 2,0,8,9,6,6,1,1,
+ 3,0,10,10,6,6,1,1,
+ 4,0,6,9,14,11,2,2,
+ 5,0,9,10,6,6,2,1,
+ 6,0,8,10,17,13,2,2,
+ 7,0,8,10,10,8,2,2,
+ 8,0,9,10,6,6,1,1,
+ 9,0,6,10,9,7,2,2,
+ 10,0,6,10,12,14,2,1,
+ 11,0,2,6,15,8,3,2,
+ 12,0,0,8,11,11,2,2,
+ 13,0,8,10,6,6,2,1,
+ 14,0,3,10,8,8,1,2,
+ 15,0,10,10,7,6,2,2,
+ 16,0,7,10,6,6,2,3,
+ 17,0,6,9,6,6,1,2,
+ 18,0,4,9,9,10,2,2,
+ 19,0,7,10,15,14,2,2,
+ 20,0,9,9,13,8,2,2,
+ 21,0,2,10,7,15,2,2,
+ 22,0,5,9,9,9,3,3,
+ 23,0,9,9,12,12,1,1,
+ 24,1,5,9,17,15,3,3,
+ 25,1,4,6,6,9,1,1,
+ 26,1,8,10,9,8,2,2,
+ 27,1,7,9,6,13,1,2,
+ 28,1,8,10,7,7,2,2,
+ 29,1,7,10,9,7,2,2,
+ 30,1,10,10,6,6,2,2,
+ 31,1,0,9,14,14,2,2,
+ 32,1,3,10,12,11,1,2,
+ 33,1,9,10,13,14,3,3,
+ 34,1,6,10,11,9,2,1,
+ 35,1,5,9,12,6,2,2,
+ 36,1,0,4,8,8,1,1,
+ 37,1,6,9,7,7,1,2,
+ 38,1,8,10,6,6,3,2,
+ 39,1,9,10,6,6,2,2,
+ 40,1,3,9,6,7,1,2,
+ 41,1,4,10,8,10,1,1,
+ 42,1,7,7,9,9,1,1,
+ 43,1,7,10,13,14,2,2,
+ 44,1,9,10,6,6,1,1,
+ 45,1,7,10,8,11,2,2,
+ 46,1,9,10,11,6,2,2,
+ 47,1,9,10,7,6,2,2,
+ 48,1,6,8,6,6,1,1,
+ 49,0,10,10,6,6,2,2,
+ 50,0,8,10,6,6,2,3,
+ 51,0,4,9,9,9,2,2,
+ 52,0,6,9,6,6,1,1,
+ 53,0,8,9,6,6,2,2,
+ 54,1,8,10,11,13,2,3),
+ ncol = 8, byrow = TRUE)
> colnames(mydata) <- c("id", "treatment", "pre_knowledge", "post_knowledge", "pre_STAI6", "post_STAI6", "pre_Rperc", "post_Rperc")
> mydata <- data.frame(mydata)
> mydata$treatment <- as.factor(mydata$treatment)
> ## ancova analysis with CI
> lm_knowledge <- lm(post_knowledge ~ pre_knowledge + treatment, data=mydata)
> summary(lm_knowledge)
Call:
lm(formula = post_knowledge ~ pre_knowledge + treatment, data = mydata)
Residuals:
Min 1Q Median 3Q Max
-3.6683 -0.2656 0.1572 0.4593 1.6673
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 7.8495 0.3984 19.703 < 2e-16 ***
pre_knowledge 0.2416 0.0533 4.533 3.54e-05 ***
treatment1 -0.1812 0.2789 -0.650 0.519
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 1.023 on 51 degrees of freedom
Multiple R-squared: 0.2943, Adjusted R-squared: 0.2666
F-statistic: 10.63 on 2 and 51 DF, p-value: 0.000138
> confint(lm_knowledge)
2.5 % 97.5 %
(Intercept) 7.0496786 8.6493112
pre_knowledge 0.1346092 0.3486055
treatment1 -0.7411105 0.3787666
> lm_STAI6 <- lm(post_STAI6 ~ pre_STAI6 + treatment, data=mydata)
> summary(lm_STAI6)
Call:
lm(formula = post_STAI6 ~ pre_STAI6 + treatment, data = mydata)
Residuals:
Min 1Q Median 3Q Max
-4.916 -1.084 -0.668 1.346 7.693
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 2.83526 0.97016 2.922 0.00517 **
pre_STAI6 0.63879 0.09645 6.623 2.14e-08 ***
treatment1 0.41567 0.61435 0.677 0.50171
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 2.256 on 51 degrees of freedom
Multiple R-squared: 0.4647, Adjusted R-squared: 0.4437
F-statistic: 22.14 on 2 and 51 DF, p-value: 1.2e-07
> confint(lm_STAI6)
2.5 % 97.5 %
(Intercept) 0.8875916 4.7829275
pre_STAI6 0.4451564 0.8324146
treatment1 -0.8176843 1.6490261
> lm_Rperc <- lm(post_Rperc ~ pre_Rperc + treatment, data=mydata)
> summary(lm_Rperc)
Call:
lm(formula = post_Rperc ~ pre_Rperc + treatment, data = mydata)
Residuals:
Min 1Q Median 3Q Max
-1.01465 -0.30735 -0.01465 0.06681 1.06681
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.68151 0.22104 3.083 0.0033 **
pre_Rperc 0.62584 0.11047 5.665 6.82e-07 ***
treatment1 0.08146 0.13228 0.616 0.5408
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.4843 on 51 degrees of freedom
Multiple R-squared: 0.3865, Adjusted R-squared: 0.3625
F-statistic: 16.07 on 2 and 51 DF, p-value: 3.88e-06
> confint(lm_Rperc)
2.5 % 97.5 %
(Intercept) 0.2377416 1.1252711
pre_Rperc 0.4040715 0.8476078
treatment1 -0.1841094 0.3470364
> #### results