A 300 Year History of Text Analysis

February 15, 2010 | No Comments

Social media research is this year’s big new thing. Researchers are all over sentiment analysis, text analysis, and content analysis heralding it as the newest and greatest thing to hit marketing research. You may be surprised to learn that content analysis is, in fact, older than my great, great, great, great grandmother. History buffs will enjoy this very short history of content analysis.

17th Century: Religious dissent: Text analysis and sentiment analysis were used by religious authorities to identify people who seemed to be using printing presses to share information that wasn’t related to the church.

1890s: Newspaper analysis digs in: In 1893, one of the very first content analysis of newspapers was published! This particular study attempted to demonstrate how publications were shifting their focus from religious and science topics to gossip and scandal (sound familiar?).

1930s: Newspaper analysis settles in: Before the internet and computers, researchers used to buy a copy of every newspaper and manually cut out articles on the topics they were interested. Someone might be tasked with cutting out all the articles about the current election. They would then manually analyze every article to identify the mood of the writer and the features of the article. With the changing economic climate, the 1920s and 1930s led to huge growth in content and text analysis. People felt the mass media were at least partly responsible for all the woes of the world. This was a great era for the development of both survey and opinion research, which most people are familiar with, as well as content analysis, which has mostly remained of interest to qualitative experts.

1970s: Survey research: Survey research really takes hold. You already know about verbatims and open-ends on survey questions. Multiple choice questions often have a place for people to write in an answer if they feel the question didn’t already include their answer. Those verbatims go through manual sentiment analysis and text analysis whereby people read each and every verbatim to analyze the mood of the writer and the features of the verbatim.

2010s: Social media research: And we arrive at today. We may not being doing text analysis manually anymore, but with one hundred years of solid research and learnings, the research community has amassed a lot of skill and experience.

If you are interested in reading more, Klaus Krippendorff, the leading expert in this field, has written many fabulous texts about content analysis. It’s a tough read, but a great read.

No Responses to  “A 300 Year History of Text Analysis”