101 Days of Community Psychology

I’m a community psychologist by training/degree, though like most quasi-academics, my portfolio and interests aren’t constrained to one field. This is the reality for community psychology. It is extremely broad from a content and methodological standpoint. However, it’s really important to stay on top of recently developments and trends, so I wanted to take a look at what’s been published recently.

In the spirit of recent posts on 101 Days of Evaluation and 101 Days of Implementation Science, let’s take a look at the last 101 Days of Community Psychology. The SCRA webpage has a really nice list of journals that focus on community psychology. I used that to determine what journals to pull from, zeroing in on the list of 12 below. I also added links to these so you could check them out yourself.

  1. American Journal of Community Psychology
  2. Journal of Community Psychology
  3. Action Research
  4. American Journal of Orthopsychiatry
  5. American Journal of Health Promotion
  6. American Journal of Preventive Medicine
  7. American Journal of Public Health
  8. Global Journal of Community Psychology Practice
  9. Australian Community Psychologist
  10. Health Education and Behavior
  11. Health Promotion and Practice
  12. Journal of Primary Prevention

Using a Pubtrawlr interface, I pulled all their abstracts from the last 101 days. This left me with 670 unique articles. A brief tangent: that’s a lot of articles. To keep on top of the field, you would have to read nearly seven articles a day. The most disciplined researcher/practitioner I know has made it her personal quest to read and take notes on ONE article a day. I’m in awe. Nonetheless, that’s a lot of articles to stay on top of, even if it just involved skimming the titles.

To make this more efficient I used some NLP techniques. As part of high-level overview, I like to look at word pairs. This shows us a lot of articles centered around health. Some of the specific topics that seem to be coming up include COVID-19 (which is great that articles are getting reviewed and out), physical activities, food insecurity, and low income. There’s a few other pairs that clearly refer to some of the methodological elements like PsycInfo –> database –> record.

The size of the node corresponds to number of mentions and the width of the arrow corresponds to number of connections. I have an interactive version of this graph, but haven’t yet figured out how to get those types of files in WordPress. Sorry!

However, we can take a more refined look using LDA models. I’ve written about these in some previous posts. We see articles centered around opiate treatment (Topic 10), cancer screening (Topic 11), and smoking (Topic 18). When I run LDAs, there are sometimes “garbage collector” topics that pick up leftover words, but this looks pretty clean. There are few metrics that show the overall quality of the topics, though that’s a little outside the scope of this post. Check out Boyd-Graber et al (2014) for more information.

But we can take it even a step further. We can see how all the 670 articles relate to one another. Below is a UMAP clustering graph, where each point represents an article and each color corresponds to the above topics. The cluster all the way off at the right is Topic 6.

What UMAP does it take a multidimensional matrix and reduces it down to facilitate visualization.

This analysis gives us a sense of what’s being covered recently in the field of community psychology. What I take away: Community Psychology is indeed broad. The topic models show clusters around different problem domains. That said, we see a focus on health and developmental issues, with several topics focus on children, youth, or adolescents. This is consistent with the preventative focus of the field.

We also (luckily) see a bunch of responsive journal editors. As of this writing mid-October 2020, the COVID crisis is about seven months old. Despite that novelty and the long process of academic publishing, we see a number of articles focused on COVID 19. This is good. We want to be responsive to the current needs of our communities, and it’s super that those needs and associated research questions are already filtering up through academia.

This process of retrieval –> synthesis –> reporting is what PubTrawlr is all out. PubTrawlr is a web-based application to help practitioners and researcher make sense of large amount of scientific articles more effectively and efficiently. Soon, the website will be moving into beta testing. with a full, open launch early next year. For more information, feel free to reach out to us at info@pubtrawlr.com and follow us on Twitter @Pubtrawlr.