Research Center Blog

This next set of posts, the second in our six-part research series, will answer three questions:

1. What factors matter most when you want to retain your best staff?
2. What factors matter most when you want to ensure parents are highly satisfied?
3. What factors matter most when you want to ensure students are highly engaged?

Earlier in September, we discussed which student survey items scored the highest and lowest and which items increased and decreased the most year-over-year. (If you missed that one, start here.)

When we present data like that to a school district, we almost always hear, “This is really valuable. Where should we begin? What should we fix first?”

This blog answers that for you as it relates to students. (To see the staff “what should you fix first” post, click here. For parents, click here.)

What affects whether a student enjoys being at their school?

There may be some common, perceived reasons students like their school or not.

• Enrollment (maybe if I feel like I’m getting lost at school and I’m just the number on my ID?)
• Student-teacher ratio (perhaps there are so many kids in my algebra class that I can never get help when I need it?)
• Achievement (maybe I feel a lot of pressure and stress in every class just to get into a good college? Or maybe I don’t feel like I’m being pushed enough?)
• Salary (students won’t know this, but they can tell whether their teachers like their jobs or not!)
• Community wealth (certainly students know whether they and other students aren’t eating and are hungry all day)
• Locale (is there a perception that smaller schools are “better” than large urban ones?)

Like we said in our staff and parent posts, maybe these factors affect students’ satisfaction and maybe they don’t, but here’s the problem: you don’t have much (if any) control over these things, especially in the short term. You can’t change the enrollment of your schools. You aren’t going to move your school to the suburbs. 

The good news, though, is that these uncontrollable factors above only explain about 18% of the “how much students enjoy school” variance. In other words, these factors explain students’ satisfaction much worse than they do for parents and staff.

So, while these factors matter a little, there are lot of things you can do to improve retention that are very much within your control. (Stats a little rusty? Miss our explainer in the staff and parent posts? See the R2 explainer in the box below. Otherwise, skip it!)

Variance Explained? Huh?   Let’s say I tell you that I can predict exactly how many touchdown passes Jordan Love) will throw this year and that the only data point I need is the number of hours he practiced this offseason. If I’m correct year after year, I’ve explained 100% of the variance (R2 = 100%). I don’t need to know any other piece of information to make a perfect prediction. Take the other extreme. Maybe practice hours have absolutely nothing to do with how many TDs he throws. In that case, practicing explains 0% of the variance (R2 = 0%). In reality, the time spent practicing is one of many things that affect TD passes. Practice might explain, say, 15% of the variance, but to make near-perfect predictions, I’d want to know a bunch of other things: how good his receivers are, what the weather is for each game, if he (and his receivers) is healthy, how good our opponents are, and so on. The goal is simplicity: get as close to 100% as possible with as few variables as possible.

What should you fix that is within your control?

This gets us back to our initial question: If you get some scores back that aren’t what you want them to be, what should you fix first to ensure students enjoy being at their school? 

• I know my teachers care about me.
• I can be myself at school.
• Most kids at school follow the rules.
• If I were bullied, I would feel comfortable talking to someone about it.
• I like helping other people when I can.
• I talk to people outside of school about what I’m learning.
• I can relate to what I’m learning at school.
• My teachers make their classes fun and interesting.

Note the very clear theme of interpersonal relationships. Students need to know they are cared for, that they are safe, and that adults are engaged in what they’re doing at school – both staff and parents.

Revisiting our R2 metric above, together, the survey items listed above explain 89% of the “how much students enjoy school” variance. In other words, with just a few questions for students, we can almost perfectly predict whether they enjoy being at school.

 

What’s a “good” R2?   Human behavior is notoriously difficult to predict. We’re complicated creatures. We’re all a little strange, illogical, irrational, and inefficient. Thus, in social science research, if we can explain 50% of the variance (R2 = 0.50), we’re in great shape. The fact that we reached nearly 90% (R2 = 0.90) is thrilling for us nerdy data folks.

Coming up, one of our project managers, Scott Girard, will tell you who’s doing great work in these most-impactful areas, and, more importantly, how? What, exactly, are they doing? That’s the practitioner side of things. At the same time, our Senior Research Director, Derek Gottlieb, will couple Scott’s work with what peer-reviewed research is saying about what you need to do to keep your best staff.

---

The School Perceptions Blog and Resource Center features the voices of our team members. This post was written by Rob DeMeuse, Vice President of Research.