Correspondence about validity of the Adjective Checklist

Below is a slightly edited correspondence that I had in the summer of 2006 that I’ve been meaning to post ever since. Jack Arnold is a retired professor of psychology in California. After doing the study that is linked in his first email message to us, he applied his technique using the Adjective Checklist from this website. He found that our adjectives are representative enough of personality traits in general that they can be used to identify “types” of people (see Figure 1 below) as those people are described on the internet.

Rod Paynter

17/08/06

Jack Arnold to info@calsca.com

Dear Group,

I found your Adjective Checklist in the course of a search for material to follow up my recently published article in Current Research in Social Psychology (http://www.uiowa.edu/~grpproc/crisp/crisp11_12.pdf). The thrust of the paper was that Google page counts for the occurrence/co-occurrence of search terms contained semantic information that could be "mapped" using cluster analysis and/or multidimensional scaling.

Recently, it occurred to me that some of the same kind of semantic information might be extracted using a more conventional approach--a kind of Google Q factor analysis. I obtained co-occurrence counts for the names of 20 famous persons and 207 of the adjectives from your list. I then did a principle components factoring of the correlations among the 20 names. The results were impressive. The first principle component was not especially interesting. It appeared to reflect the degree to which the "famous name" profiles were affected by the relative Google frequencies of the adjectives. The second and third principle components were, however, very interesting. The plot of the names against their loadings on these two components showed several distinct and easily interpretable clusters. (Note that "No Name" is a dummy variable that simply reflects the unpaired Google occurrences of the adjectives.) I have attached a figure depicting these results. I hope that it makes it past your email filters. The figure is embedded in an MS Word page.

My reasons for directing your attention to this matter are two: First, it is clearly a cheap validity study that supports your Adjective Checklist. I hope that you find that encouraging. And, second, I am curious about the list's provenance. How did it come to be and how is it related to some of the better known--but less accessible--lists. I am unlikely to do any more systematic work on this particular problem, but if I can find someone who will (do you know someone?), they are likely going to need that kind of information.

I hope you find this interesting and can help.

Best wishes,

Jack B. Arnold

Professor of Psychology, Retired

Saint Mary's college of California

19/08/06

Rod Paynter to Jack Arnold

Jack, this is very interesting! In my layman's words, is it accurate to say that you did Google searches with combinations of names and adjectives, and found that certain "types" of people were correlated with distinct groups of adjectives?

The adjective checklist that you found on the CALSCA website is used in Life Skills programming as a stimulus in a feedback/group building exercise. It began as a much smaller list, and has been occasionally added to by interested folk.

May I put the figure and your explanation on our website? And a link to your paper?

Regards,

Rod Paynter

19/08/06

Jack Arnold to Rod Paynter

Dear Rod,

Thanks for the additional information on the adjective checklist.

I would be delighted to have the figure appear on your website. Your interpretation of what the data show is, I think, essentially right. The factors or components are essentially summaries of how the adjectives describe the personalities. Just which adjectives figure how much in the determination of which factors, I have not yet determined.

If you know of someone who wants my data, let me know. I'd be happy to send along my rather large spreadsheet. I still enjoy fooling around with data, but I'd rather be helping my five-year old granddaughter pull weeds and inspect bugs.

Again the best,

Jack