We are happy to report that our latest efforts at collecting data are bearing fruit. A few of us (we won't say who) are rather addicted to checking Qualtrics every day to see how many responses we have to our survey. After all the time we've put into creating and revising both our methodology and our survey, it's really exciting to see the hard work paying off.
Our biggest focus right now is on power. There's a lot to say about classism and power and we'll get into that eventually, but right now we're focused on a different type of power: statistical power. If you've taken a research class before, you may remember that statistical power refers to the ability to reliably detect an effect if one exists. If our sample size is too small then we won't have enough statistical power to draw any reliable conclusions from the data we collect. Any differences we see between groups in our analyses could simply be due to chance. So, we needed to figure out how big our sample must be in order to have enough statistical power to confidently state that any differences we observe are not simply due to chance.
Figuring out our desired sample size means knowing how big our population of interest is. In other words, we need to know how many people are enrolled in a counseling program at either the master's or doctoral level in any given year. That's not easy to find out. We could use the CACREP Vital Stats reports, but those only include enrollment in CACREP-accredited programs. We are trying to survey students enrolled in any counseling program regardless of accreditation status. Still, a quick review of the CACREP Vital Statistics reports tells us that, on average, there are about 40,000-45,000 students enrolled in CACREP-accreditated counseling training programs every year. That's actually helpful information because even though that number doesn't include students enrolled in non-CACREP programs once your population of interest hits around 50,000 people the minimum number of responses you need to have to believe that your sample is reasonably representative of the population doesn't change much - it just depends on what you want your margin of error to be. We'll spare you the rest of the statistical jibber-jabber, but right now we'd like a sample size of at least 385 respondents to have the kind of statistical power and margin of error we're looking for. There will still be limitations to the conclusions we can draw from our data based the simple fact that our methodology relies on a convenience sample instead of random selection of participants, but getting to 385 individual responses will give us a confidence boost (literally and figuratively). The good news is that we're close. But the bad news is that we aren't there yet. So, we're going to do one more round of recruitment and hope that we get enough responses to put us over the finish line.
You can help! If you're a student or faculty member, you are free to forward information about our study to any other student or anyone who has graduated from a counseling program in the last 5 years. If you're a supervisor or counselor in practice, you can forward information about our study to any current student you know or any other counselor who has graduated in the last 5 years. You can send them a link to our study page, send them the direct link to the study at https://tinyurl.com/PrivilegeinCE or just forward the Canva image below:
Comments