This post is part of a series inviting discussion, comments and reflection on the results of my dissertation.
Now we start looking at the results of my study. Every variable was examined to see its relationship with how often the school was using curriculum videoconferencing and whether that variable could be used to predict the use of videoconferencing. This is why my questions about the method of measuring utilization are important.
In this post, we’ll look at the relationship between the school’s use of curriculum videoconferencing and some school demographic variables.
- The elementary schools use videoconferencing significantly more than the average of the other levels (r=.280, p=.000).
- The high schools use videoconferencing significantly less than the average of the other levels (r=-.194, p=.001).
- Where the coordinators support all levels, their schools are using it significantly less than the average of all the other levels (r=-.202, p=.001).
- Population is not significantly correlated to the school’s use of videoconferencing. (However, when using multiple regression in analyzing school demographics, population contributed a very small amount to the prediction model. Schools with higher population were using videoconferencing slightly less (b=-8.03, p=.021).)
- Poverty scores (National School Lunch Program, US only) are not significantly correlated to the school’s use of videoconferencing.
- The only ethnicity that is using videoconferencing significantly more than the others is the “other” category (r=.202, p=.001). Most of the coordinators who chose the other category wrote in Native American or Canadian First Nations (16 out of 22).
- If faced with a choice, install videoconferencing in elementary schools first.
- If you coordinate videoconferencing for multiple levels/schools, you need to train and support a VC coordinator in each school.
- Curriculum videoconferencing meets a particular need and is often used by Native American and Canadian First Nations schools with access. This should be supported & continued.
Stats Principles to Remember
- Correlation doesn’t necessarily mean causation. You cannot say with confidence that because the school was an elementary school; that caused the school to use videoconferencing more.
- r is the measure of the correlation. In medicine, we want a very high r (close to 1 as possible); in education and other social sciences, we’re doing good to have it close to .5.
- p tells you how much of that result is due to chance (for example .001 is .1% due to chance)
- Do you agree with the results? Would you recommend installing videoconferencing systems (with the purpose of curriculum videoconferencing) in elementary schools first? Is that where VC is available in your area? Why or why not?
- Do you think that it is important to have a VC champion/coordinator in each school?
- This is my first results post. Is it clear? Does it raise questions? Do you recommend another way?