Explain It Like I’m Five: Equity, Community Psychology, and NLP

Spread the love

Community Psychology is a misunderstood field. It’s not people who sit on couches talking about problems. Instead, community psychology looks to build people, organizations, and systems up. We want to further health and wellness (broadly defined) in the places in which people live. Because of that, community psychology is deeply rooted in a common value system; one that includes social justice and health equity.

We here about PubTrawlr recently published a synthesis of how community psychology talks about equity in some of the keys journals. Because we’re all about research synthesis here, I used a few common natural language processing methods. Of course, how would I explain this to my soon-to-be-kindergartener? Let’s give it a go.

There are lots of articles published; too many to keep up with. But community psychologists need to stay on top of these articles to make sure we know the best ways to work with people in communities. We want to make sure that we work with everyone. This is called equity; making sure that we are working with the people who needed it.

Community psychologists need to know about equity. Natural language processing can help us to read lots of articles really fast. In this article, we looked at all the articles in five different journals that community psychologists sometimes read. We thought that there would be a ton of articles, too many to read.

But it wasn’t. I only found 155 articles. Only about 1 in 7 articles had any mention of equity. That was surprising. Here are some of the words that showed up across all the articles.

I then looked at the different topics that people wrote about. This is shown in the figure below. A lot of the articles were about “community-based participatory research.” This is a type of approach that makes sure to include the people being studied into conversations about the research. We community psychologists love CBPR. It showed up in about 14 articles!

How often different topics occurred across all articles.

I also looked at how words are related to others. All scientific fields have their own special language, called “jargon.” Jargon can make it difficult to explain findings to others. The below figure showed how different the top 35 words used by authors were related to others, meaning that these words were sometimes used close to one other. Strangely, we didn’t see “equity” being closely related to “disparities.”

Something that’s not for a 5-year old. This is called a UMAP projection. Basically, I vectorized all the words, and then reduced the dimensions from 128 to 2 so that I could plot them in a graph that preserved some of the global relationships.

There was one last step. I found the article that is most like each topic. I put these recommended articles into a table that can be downloaded below. Here’s a snippet of it. That philosophy article looks pretty interesting!

Just a few of the recommended articles.

Read the article here.

If you want the full article and are having trouble accessing it, reach out to me at jon [at] pubtrawlr.com, and I’ll be delighted to send it to you.

Leave a Reply

Your email address will not be published. Required fields are marked *