“Personality tests should not be used for personnel selection”

The suggested reading that everybody should have read.

Morgeson et al. (2007b)

Morgeson et al. (2007a)

Additional reading, that we do not expect people have read.

Fleeson (2004)

Chen (2018)

We will use the papers as inspiration of the slides, but we will not go into the depth of those.

There exist a lot of different personality test, and common for them are that they measure “personality”.

They are based on a questionnaire with generally many that measure different facets of the person. All facets are latent traits.

The personality test measure multiple factors at the same time, where some items could measure multiple facets at the same time.

The endgoal of the personality is typically to categorize people into different personality groups. So statements like “He is a very open person” or “He is an extrovert person”.

When we have items and factors in personality questionnaire, we have learn a lot of different tools to evaluate these questionnaire.

For reliability and validity we have learn about Cronbachs \(\alpha\) and MacDonald \(\omega\) for unidimensional factors. While we have learn about CFA and IRT when looking at validity of the construct and the measurement error of these constructs.

With these tools we have a good theoretical foundation to talk about personality test and the consequences of the use of categorization in personality.

Furthermore we have in linear models learn about inference to generalization.

The items in the questionnaires are often a what will you do in this scenario with quite limited responses categories.

The way it is constructed, there is no right or wrong answer.

It is only test result that other persons will use. So the respondent could fake their response to achieve what they think other people what to hear. In this sense people in their test results could be different from their true personality, Morgeson et al. (2007b) and Morgeson et al. (2007a) .

The first paper Morgeson et al. (2007b) was conducted as summarization from a table discussion with chief editors of a journal: Personnel Psychology and the Journal of Applied Psychology. Each editors gives their own personal summarization of each question in the table discussion.

They assume there is a dependency of personality type and job-success. However they do not specify the success-criteria, and they briefly talk about personality types and not much about personality tests.

They spend much time about validity, but it is in general terms, and not specific about a personality test. They speak about validity broadly, however validity is only about single test.

Morgeson et al. (2007b) are concerned about faking could affect the validity of the test-construct.

Since the Morgeson et al. (2007b) was a talking paper with a lot of topics and unspecific in their concerns, there have been a lot of critic, which Morgeson et al. (2007a) summarize and answer with more specification of their meaning.

The summarization are giving in the abstract of Morgeson et al. (2007a): “(1) Our criticisms of personality testing apply only to the selection context, not to all research on personality; (2) the observed validities of personality tests predicting job performance criteria are low and have not changed much over time; (3) when evaluating the usefulness of using personality tests to select applicants, one must not ignore the observed, uncorrected validity; and (4) when discussing the value of personality tests for selection contexts, the most important criteria are those that reflect job performance. Implications.” Morgeson et al. (2007a, 1029)

The most famous personality test are the big 5, (OCEAN), which try to measure different attributions of people.

• O = Openness
• C = Conscientiousness
• E = Extroversion
• A = Agreeableness
• N = Neuroticism

The facets are doupols, since if you are not extrovert then you are intervert.

The results are on the facets of the OCEAN test, but people are grouped together based on their OCEAN test.

Ideally all combinations of OCEAN should exist.

The OCEAN are often used in a work-related issues, such as recruitement, categorization of workers, etc.

Another famous personality, whom have gain popularity in the recent years, are the 16 personality test, where people are classified into 16 personality types, based on 4 facets.

• Introversion/Extraversion
• Sensing/Intuition
• Thinking/Feeling
• Judging/Perceiving

This personality puts you into either group of the facet.

The result is that a people can be categorize in groups such as, ISTJ, ESFJ, EIFP, etc, where is groups are summarized by the type.

The 16 personalities are used in dating apps, such as Tinder.

We have constructed 4 groups, which will get a specific aspect of the topic to discuss. The plan is that the constructed groups will talk in 15 minutes about their specific aspect, where you will find arguments for and against the statement. We have selected one of each group to present orally the viewpoints of the group.

To have debate every group will get 5 minutes to discuss each specifik aspects in generelle terms. The idea is that you will come with arguments supporting or against the presenting group. This is alive the debate.

The group orally present, the other groups add, and Magnus & Morten will be the advocate of the devil.

• 15 minutes to discuss prepare your specific aspect
• 5 minutes to discuss the second aspect
• 5 minutes to discuss the third aspect
• 5 minutes to discuss the last aspect
• 15 minutes Break
• Debate of aspects, 10 per aspects.

Statement: “Is personality something unique, or something you can create from constructs?”

Statement: “How much can you trust the categorization of the person in a construct? And when do we trust it?”

Statement: “Can you use the results in a personality test as a screening tool in dating? Would you reject a person based on their personality testresult?

Statement: “What matter most at the job: The skill of the person, or the claimed personality of the person?”