To my understanding, the main reason that Great Colleges survey is not readily available is because we had the climate survey. Simply put, there is no reason for two different surveys.
To my understanding, the argument should not be towards administering or sharing the survey (we all agree on that) but rather on which survey of the two will provide us with “meaningful, assessable and verifiable data ”
I have taught Statistics here at the college since fall of 2011, I have been with the Data Team for about as long, and I have taken two courses via Achieving the Dream on designing IR research (and did very well in both). Also I have taken stats courses in my previous life, when I was a college student both in the States and in Greece. And I have seen both surveys, and my professional opinion on the matter is that the Great Colleges survey is the better survey. It is designed by statisticians that have forgotten more statistics than we will ever learn, the questions are specific, and do not skew responses in one direction or another. They are clear, unambiguous and unbiased. Also it allows us to compare ourselves with other colleges nationwide. …show more content…
Simply put, the climate survey allows us to collect longitudinal data the Great Colleges survey will allows us to collect both longitudinal data and latitudinal data (by comparing our population at single point in time with other colleges).
Or in other words, suppose you weigh yourself every morning, then over time you can tell whether you are gaining or losing weight but not whether you are too underweight or overweight. Only if you compare yourself to other people you can tell if you need to eat more or less. As a sociologist would argue, we need to put things into
perspective.
I want to be honest, if we were to go with our climate survey will need to review the questions for possible biases and ambiguities. Also we will need to clean, code and put our data in a flat file. Then we will need to create a codebook. It is important to note that open-ended questions and/or responses will take a lot of time and effort to code. Because data needs to be standardized for data analysis and open-ended questions will need to be coded into nominal categories.
I have to caution you that this decision should not be done as a “popularity contest”, but rather in a scientific way where the people that have the skills and expertise review both surveys and offer their opinions.
Regardless the decision we reach, I am committing that as long as I am the chair of the Data Team, we will keep collecting, analyzing, and sharing the results of the survey we choose to conduct.
Have a wonderful Easter weekend!