Versions of IPIP Measures around the World Wide Web


Because the IPIP is in the public domain, anybody can distribute paper-and-pencil copies of IPIP measures or administer online versions of IPIP measures without asking for permission or guidance from the researchers who created and maintain the IPIP website.

In some cases, others have asked for our assistance and guidance in implementing IPIP materials, but in many cases, they have simply used IPIP materials on their own. In the latter cases, we cannot guarantee the quality and validity of what others have done with the IPIP.

High-Quality Implementations of IPIP Measures

We are aware of some excellent implementations of the IPIP. For example, Jordan B. Peterson, who co-authored an article on the development of the 100-item IPIP 10 Big-Five Aspects Scales (BFAS; DeYoung, Quilty, & Peterson, (2007), has created, with Daniel M. Higgins and Robert O. Pihl, a website where you can complete the scales online and receive a narrative report describing your personality. The site, Understand Myself, has helped thousands of individuals make informed choices about their education, career, and life in general. The cost for taking the BFAS, $9.95, seems reasonable for this data-driven, evidence-based online personality assessment.

The Open-Source Psychometrics Project has implemented the IPIP Big-Five Factor Marker (BFFM) scales into an online inventory. The 50-item BFFM inventory is probably the most widely-used of all IPIP measures because it has been used as a sample questionnaire for measuring the five major personality factors since the IPIP website first came online in 1998. Over a million persons have completed the Open-Source Psychometrics Project version of the BFFM inventory.

The IPIP consultant  (Johnson, 2014) developed online versions of the 120-item and 300-item IPIP representations of the NEO PI-R (Costa & McCrae, 1992). Over a million persons have completed either the shorter or longer version of the IPIP-NEO, which has been online since 1996. Because Johnson has freely distributed the computer code for online administration of the IPIP-NEO, numerous individuals have used the code for creating their own online versions of the IPIP-NEO. To the extent that these copycat sites have faithfully used Johnson's code, they should produce the same results as the original online IPIP-NEO. It has not been feasible for Johnson to locate and quality-check all of these other sites, so we cannot vouch for their accuracy. However, one of the best online implementations of the 120-item IPIP-NEO can be completed at no cost at https://bigfive-test.com/ . This implementation is available in 18 different languages and has been completed by over 4 million persons. If you are a programmer who is interested in the code for this site, it is freely available at GitHub.

One Problematic version of the IPIP-BFFM Scales

Unfortunately, because the IPIP is in the public domain and developers can do anything they want with the IPIP, sometimes problematic versions of IPIP measures have appeared on the World Wide Web. Perhaps the most troublesome of such measures has been an early implementation of the Big-Five Factor Marker (BFFM) scales. This implementation appeared in 2011 by the Open-Source Psychometrics Projects, which at that time was called personality-testing.info (that URL is no longer in service.) While we list the current version of the BFFM scales at the Open-Source Psychometric Projects among the high-quality implementations of IPIP Measures, the original version at personality-testing.info contained a number of serious problems that have confused countless users. That problematic version of the BFFM scales can still be found in PDF form on the Open-Source Psychometric Projects website. Because the PDF lists the IPIP home page at the top of the document, we frequently get inquiries about the scoring and interpretation of what we now call the "rogue version" of the BFFM scales. One reason that we created the current web page was to describe three major problems with this rogue version.

The first problem is that the rogue version uses scale labels from Costa & McCrae's (1992) Five-Factor Model instead of the lexical Big-Five Model (Goldberg, 1992). This means that the fourth factor of the Big-Five model, Emotional Stability, is labeled Neuroticism (which is the opposite of Emotional Stability), and the fifth factor, Intellect, is labeled Openness to Experience. One can work around simply by using the proper labels. Note that the formula in the PDF for scoring factor 4 actually produces scores in the direction of Emotional Stability instead of Neuroticism, despite the score being labeled "N" for Neuroticism. (Follow this link for a more detailed explanation of the differences between the Five-Factor Model and the Big Five, under the heading "Finding a Multi-Scale Personality Inventory.")

The second problem is that the rogue version uses an atypical scoring system for IPIP measures. The standard scoring procedure for IPIP scales is to code +keyed item responses from 1 to 5 and -keyed item responses from 5 to 1, and then sum all of the item responses, as explained at https://ipip.ori.org/newScoringInstructions.htm . The  standard scoring key for the BFFM scales is here: https://ipip.ori.org/newBigFive5broadKey.htm . But instead of recoding the -keyed items and summing all of the items responses, the rogue version sums the +keyed items, subtracts the -keyed items, and adds a constant. The constant that appears in the five scoring formulas in the PDF is equal to (6 x the number of -keyed items) - 10. So, for example, the first equation for E has 5 -keyed items, so the constant is (6x5)-10 = 20. The second equation for A has 4 -keyed items, so (6x4)-10 = 14, and so forth. Multiplying the number of -keyed items by 6 is mathematically the same as scoring -keyed items from 5 to 1 and adding them instead of subtracting them. The reason the rogue version subtracts 10 from the constant is that this makes scores range from 0 to 40 instead of the usual 10 to 50 with the standard IPIP scoring method.

The third problem is that the graph at the bottom of the PDF showing the means and distributions of scores uses mean item scores (on a 1-5 scale) rather than scores based on the sum of 10 items. For example, someone who scored a 40 out of a possible 50 on Intellect (which is a 30 out of a possible 40 on Openness if you use the rogue formula) would be represented by a mean item score of 40 ÷ 10 items or 4. That is why all of the scores in the graph range from 1 to 5.

Needless to say, this rogue version has confused countless people. That is why we recommend using the scoring keys and procedures on the IPIP website instead of the formulas in the PDF of the rogue version.

References

Costa, P. T., Jr., & McCrae, R. R. (1992). Revised NEO personality inventory (NEO PI-R™) and NEO five-factor inventory (NEO-FFI): Professional manual. Odessa, FL: Psychological Assessment Resources.

DeYoung, C. G., Quilty, L. C., & Peterson, J. B. (2007). Between facets and domains: 10 aspects of the Big Five. Journal of Personality and Social Psychology, 93, 880-896. https://doi.org/10.1037/0022-3514.93.5.880

Goldberg, L. R. (1992). The development of markers for the Big-Five factor structure. Psychological Assessment, 4, 26-42. https://doi.org/10.1037/1040-3590.4.1.26

Johnson, J. A. (2014). Measuring thirty facets of the five factor model with a 120-item public domain inventory: Development of the IPIP-NEO-120. Journal of Research in Personality, 51, 78-89. https://doi.org/10.1016/j.jrp.2014.05.003

[Page created 02-21-2022, updated 03-03-2022]

Return Home