Seriously, as soon as the book came out, we were up and running with it during our summer semester in our MSW program. We thought you might be interested in how we are using it and R in our Master's classes.
To set the stage properly, this book is not about any particular R package at all. It is really a guide on how to do program evaluations using R instead of other proprietary software. The majority of the book steps through different types of evaluations with case studies. They range from very simple evaluations where you might want to simply describe the characteristics of clients to much more complex outcome evaluations with multivariate regression models and, of course, everything in between.
In our MSW classes, we used R to do a class project where the students evaluated our educational program with the outcome being overall student satisfaction. We could go on and on describing what this project was like, but a picture is worth 1,000 words, and we thought it might be more interesting to show you some slides from one of our groups' final presentations. But first.... a shout-out to our students who agreed to share their work with us, so thank you Avallon Leopold, Ashley Moses and Alyssa Tanz!
The overall research question was what: factors contributed to student satisfaction in our MSW program. Avallon, Ashley and Alyssa, after conducting a literature review, hypothesized that fieldwork placement, supervision, and academic instruction would all be positively related to student satisfaction. They also thought there would be no significant relationship between marital status and overall student satisfaction. Let's tune in to see what they found....
With regard to fieldwork supervision:
The students found a weak, but non-significant between students' perceptions of fieldwork supervision and overall satisfaction with the MSW program.
About fieldwork satisfaction:
Nothing! Nada! No relationship at all!
When testing their hypothesis about marital status, this is what they found:
Still no significant relationships, but what what about students' satisfaction with academic instruction?
Whoa! We finally found something! A moderate and significant relationship between students' satisfaction with their instructors and with the overall program. You know what that means to us as professors? PRESSURE!
On a more serious note, this is how our students produced such fabulous work in 7 short weeks - hard work! To do those really cool scatterplots with the regression line superimposed, they used the scatterplot() function from the car package. The correlations were done using the rcorr() function from the Hmisc package because it helped the students deal with missing values and calculate the chances of making a Type I error in one easy step. The snazzy box plots and one way ANOVA were put together by using functions from base R. No fancy packages.
We asked the students at the end of the summer about their experiences with R anecdotally and this is what we heard:
- easier to use than that really, really expensive software package because all the commands were on one-line and they didn't have to go through menu after menu.
- they found the book really helpful because everything, including all the commands needed to do their analysis were in there.
- R just wasn't nearly as scary as they thought it would be!
All-in-all, a good first experience for us using Making Your Case with our MSW students. Oh, and by the way, Making Your Case is available at all the usual places you might think of: Amazon, Barnes and Noble and Oxford University Press. We asked Oxford to produce the book in paperback to keep the cost down and they did!
No comments:
Post a Comment