This is the time of the year when people either make lists of what happened last year or lists of what they think will happen next year. Since my guess may be as good as the next person's, here is my list of the things I think are going to be more popular among social science researchers based in institutions of higher learning. Since I have only 5 trends instead of the more conventional list size of 10, I am also listing 5 'things I wish would become trends.' I created the fortune cookie graphic below using Comic Life 3 to modify a public domain internet photo. 5. Qualitative meta-articles I just mentioned in my last post the "Handbook of methods for synthesizing qualitative research" by Sandelowski and Barroso published way back in 2007. It seems like just about everything that can be meta-analyzed has been, so I expect that it is about the right time for qualitative meta studies to start to become more common. 4. vignette research including factorial survey and conjoint analysisThis is not a new or novel approach. It is of course something I am very personally invested in, but the real reason I think this is going to become more popular is that the Sage Green book on factorial survey by Katrin Auspurg has just been published. There was a conference in Ireland last summer devoted to FS, and Swedish researcher Lisa Wallander has in the past few years published several very clear and comprehensive articles, including a particularly nice review of literature. The biggest issue might be for people figuring out how to create vignettes. I hope to contribute more to that discussion during 2015. 3. Partial least squares structural equation modeling (PLS-SEM)This is not the ultimate answer (I think that turned out to be '42') but is a technique that works some of the time when covariance based SEM is less appropriate. I continue to think that complex models, including HLM regression, mediator-moderator models, and multivariate approaches like SEM are the most appropriate methods for social science statistical modeling. More and more people are becoming aware of PLS-SEM, and there are now several software applications for model building. I think 2015 will be the year we will start to see it featured in areas beyond business and marketing research. 2. Mixed qualitative and quantitative research projects This 'trend' emerged several years ago, but I look for it to keep going up during 2015. I personally hope to see some more unique approaches to design and integration. 1. web-Based research softwareIt is no secret that I am a Dedoose user (and fan). I do not have the hours of experience (yet) that I do with some single license purchase programs but I am slowing moving toward nearly exclusive use of Dedoose whenever it makes sense. (I still use MS Word for a lot of first cycle coding.) Program cost, need to update (and update costs) and years of providing Windows-only products (most of the big ones finally have a Mac version - and a shout out to Hyperresearch for never abandoning Mac) gradually soured me on the license purchase products even though I still have current access to more than one. Additionally, for working with people at other institutions, the license purchase software programs rarely make sense. As far as quality differences among programs, it is my experience that people tend to advocate for whichever they were trained on (which tends to be whichever their university had a site license for, and there are some regional patterns in that). I do think there are some performance differences - and I'll go out on a limb and say that NVivo's handling of files may be better in some ways although on the minus side it creates potentially huge project files. What might set one of the big ones (including not just NVivo, but Atlas.ti and MAXQDA) apart in a meaningful way is being the first (after Dedoose) to offer a web-based low cost monthly subscription that works on any platform. But, by that time, Dedoose will be a few years ahead in that race. I realize I addressed just qualitative software; I think the same thing is going to/is happening with quan/stats software. You can already buy a limited license for some of the popular programs, but it is my understanding that PLS-GUI and SmartPLS will both be web and subscription-based. I expect others to emerge. The big companies should be watching this closely. If you have hung along with me for this long, click on 'read more' and see some trends I'd like to see in 2015. 5. No university research done on university students Unless of course you are specifically looking at an issue that has to do with university students or some aspect of the student or educational experience. Otherwise, it is clear that research on actual people who represent the population of interest has superior external validity. It might also serve as an outreach effort by the university and faculty, especially when it is possible to use participants from the community or region. Some universities seem to discourage this practice in general, but not all do. 4. More emphasis on qualitative methods in graduate social science education One doctoral program at my current institution requires 3 stats courses and 1 combined qual/quan. My Ph.D. required 0 hours of qualitative research instruction. I would like to see it closer to 50% for qual in all social science Ph.D. programs. 3. More emphasis on general research design and methods in graduate social science education I think that a lot of us completed a single methods class that was offered during the masters program (or is a masters level course in a Ph.D. program). After that, the focus tends to be on tools (usually statistical models) courses. It is a lot harder to build with just tools and no plans or blueprints; the tools classes also make a lot more sense if you have a solid background in design and methods. I'd like to see 2 semesters in any social science Ph.D. program devoted just to design/methods. 2. Academic writing instruction in undergraduate EnglishThis may be more of an issue in the south and my experience with higher ed is admittedly confined to three southern universities. But it is clear that prior training in how to write in an organized, directed, concise and intellectual style has been lacking among undergrads I have encountered. Instead, my students show more strength in writing personal reflection or one-sided persuasive/argumentative papers. I think those 'styles' should be mastered in K-12 (maybe more like K-8) with a focus on academic writing during the entire university career. 1. Use of R as the primary software in graduate statistics courses It's free. It does almost anything the typical undergrad or graduate student (or faculty member) needs to do. (If you are truly a big data person, you probably have a copy of SAS that someone gave to you, so I'm not worried about you). R works with all of the main platforms. Use of the (free version of) the R Studio graphical user interface makes R more comfortable for people who do not want to write code on a blank page. And R is something that students can use after graduation, even if they do not end up in an academic institution. I get tired of hearing about the 'steep learning curve,' all it really takes is some dedicated practice time, but so does just about everything else worth doing.
0 Comments
Leave a Reply. |
AuthorI am Sheryl L. Chatfield, Ph.D, C.T.R.S. I am a member of the faculty in the College of Public Health at Kent State University. I also Co-coordinate the Graduate Certificate in Qualitative Research and I am a member of the Design Innovation Team at Kent State. Archives
February 2024
Categories
|