Wednesday, April 24, 2019

Discussion: The Shortcoming of Modern Personality Research



The aim of this study which was conducted in the Inuit culture in northern Canada was to test the validity of the Five Factor personality model of the Big Five. The Inuit culture was chosen to test the cross-cultural validity of the Big-Five due to several reasons.

The first is that this was the first application of the Big Five model in Inuit culture, the second is that even though Inuit are located in Canada; their culture is not considered as a western culture (ITK, 2018), which is a major premise to tackle if the universality of the model was to be validated, since the Big Five was initially developed in the United States, a predominantly western culture (SD, 2019). The third is that the local Inuit language Inuktitut is exclusive to the local culture, which implies that it’s lexicon should give an accurate reflection of the local culture, therefore translating the personality items which were chosen from the International Personality Item Pool (IPIP) (IPIP, 2018), and conducting a survey in the local culture, should either reinforce or refute the validity of the Big Five model, especially since it was developed based on the lexical hypothesis, which succinctly states that the traits which are important to the local culture will eventually become part of its language (Hutson, 2013).

That said, did the results verify the universality and multicultural validity of the Big Five?

Theoretical Implications

Was the hypothesis accepted, how was it accepted, what does that mean?

In the review of the literature and when designing this research, it was hypothesized that “A five factor structure congruent to the big five will appear in Inuit culture”. Despite some limitations, which were restricted to two items in this study, items 9 which should have correlated to openness and did not, and item 16 which should have correlated to agreeableness but instead also correlated to neuroticism, were the only two items that did not meet the expectation of the literature of the research. In conclusion 23 out of the 25 items of the survey of this research loaded onto their components as expected.

Moreover the principal component analysis revealed by compressing 25 variables into representative variables associated with each trait required the researchers to run principal component analysis, which tests the assumptions to ensure reliability. All correlations and covariance were greater than 0.62 as revealed by the Anti-image Matrices except for item 9 which wasn’t accepted with a score of 0.412

The factor analysis of the other hand also revealed that as expected, the variables which the research predicted would load onto their components do so, with the exception of item 9, which did not load onto any component and item 16, which should have loaded onto one component which what agreeableness, instead loaded also on another component, which was neuroticism.

Even though it is possible that a research design error could have resulted in skewed results (Andrade, 2018), however it is unlikely. A more likely explanation for erroneous results of those two items could have been either issues on translation or unaccounted for confounding variables which influence has influenced the results. In the case of translation, item 9 and item 16 could have been mistranslated or translation and phrasing were mismatched, and item 9 is more likely to have had a translation issue since it did not load onto any of the five factors. To further demonstrate the idea, suppose the word “space” which in English can be defined as an area or a position but you intend to mean area and then translate into Arabic where the appropriate word should be “masaha / مساحة” but instead it was translated into another word which also means space “masafa / مسافة” which actually means distance would completely put the phrase out of context or render it even as a meaningless phrase. Moreover the likelihood of translation errors especially in abstract concepts, which is what the traits and facets of the big five deal with and are, occur with a higher frequency that descriptive terms of daily objects (Zhao, Solano-Flores, & Qian, 2018).

Error in the results arising from unaccounted for confounding variables is also likely, especially for item 16 which loaded onto two components of agreeableness and neuroticism. In order to further expand of the idea, the facets of those two traits were stated then compared to identify possible confounds exclusive to the local culture. Agreeableness is comprised of facets of trust, earnestness, altruism, cooperation, compliance, modestly, sympathy, and compassion. Neuroticism on the other hand is comprised of anxiety, irritability, immoderation, self-consciousness, and depression (Cooper, 2010).  There is a decent body of literature which links the neuroticism facet of self-consciousness with the agreeableness facet of modesty (Jourdy & Petot, 2017), this correlation of narrow traits could have accounted for the loading of item 16 onto both factors of agreeableness and neuroticism, and however, further research into the matter is required.

Aside from the above stated limitations, this study has demonstrated the cross-cultural validity of the Big-Five as a universal measure of personality and the international personality item pool as a universal personality item pool which has been and can be translated into most languages.

Strengths, Limitations and Future Directions

This survey was conducted on a sample size that is comprised of 1006 participants from the Inuit culture of Canada from an age group that ranges from 9 to 12 years of age. The first strength was that the sample size was large enough to produce reliable and valid results since the KMO was > 0.5 (Statistics Solutions , 2019)

Another strength was that the results were statistically significant at P = 0.00 which well above the required level of P < 0.05. The rationale behind setting the level of statistical significance at P<0.5 has to do with what the scientific community perceives as acceptable level of error occurrence (Dancey & Reidy, 2017).  There are two types of error that can occur when taking statistical significance into account. The fist is called a type 1 error, which is when the research rejects the null hypothesis when it’s true. The second is called type 2 error, which is when the researcher accepts the null hypothesis when it is wrong. When P<0.5 the probability of type 1 error is less than 5%, and when P>0.5, the probability of type 2 error is less than 5%. Therefore, 5% is chosen as a balanced probability that tolerates the occurrence of both errors (Minitab, 2017). However, in the medical field, the tolerance for the occurrence of type 1 error is 1% (p<0.01). The tolerance for type 1 error is low because human life is at stake (Dahiru, 2008). Therefore this research is in adherence with the strictest requirements of statistical significance.

            However there were some issues that may compromise some of the validity of the results, the first is that not all factors loaded onto their components, where 2 factors out of the 25 which in percentage terms accounts for 8% of the factors did not load onto the expected components. As stated prior, this error in loading could have been a result of translation error or unaccounted for confounding variables specific to the local culture.

            Another limitation is the age group of the sample size which was children between the ages of 9 and 12. Even though all the legal and ethical requirements for conducting the research was accounted for, it does not negate the fact that this age group which is comprised of minors might not be able to evaluate abstract concepts appropriately, which is what comprises the Big Five personality traits. Moreover, children do not have a fully developed personality, some psychologists prefer to address personality in childhood through temperaments, which is biologically based with some overlap with the Big Five (Twenge & Campbell, 2017).

In order to account for the limitations of this research, further personality research should be conducted in the Inuit culture regarding the local adaptation of the Big Five personality traits. Since the current research was conducted on children aged 9 to 12, some highly recommended research which should follow the current research into the adaptation of Big Five personality traits in the Inuit culture are either a longitudinal or a cross-sectional study. Those types of research are designed to study personality across the lifespan as people age and personality develops (IWH, 2015). Other research can examine the adaptation of the Big Five personality factors in an adult sample of the Inuit culture. Lastly This research was conducted using the international personality item pool, it would be ideal to compare the results with research that uses  NEO personality inventory 3, which was developed by the psychologists Paul Costa and Robert McRae , who are the psychologists who developed the Big Five personality from Eysenck’s earlier PEN model (Costa & McRae, 1995).


The research that was conducted on a sample of 1006 Inuit children aged 9 to 12 has validated the universality and cross-cultural applicability of the Big Five personality theory, by translating items from the international personality item pool into Inuktitut, which generated statistically significant results that accepted the presented hypothesis “A five factor structure congruent to the big five will appear in Inuit culture”. However there were some limitations to the research which could have risen from a failure in translation or confounding variables, were manifested by a failure to load 2 out of the 25 factors onto their respective components while conducting a factor analysis. This research’s limitation can be accounted for my conducting more personality research in Inuit culture by either using a different research design such as a longitudinal or cross-sectional study, or by using a different item pool such as the NEO personality inventory 3.



Andrade, C. (2018). Internal, external, and ecological validity in research design, conduct, and evaluation. Indian Journal of Psychological Medicine , 498-499.

Cooper, C. (2010). Narrow personality traits. In Individual Differences and Personality (pp. 78-95). London: Hodder.

Costa, P., & McRae, R. (1995). Primary traits of eysenck's p-e-n system: three- and five-factor solutions. Journal of Personality and Social Psychology, 308.

Dahiru, T. (2008). P – value, a true test of statistical significance? A cautionary note. Ann Ib Postgraduate Med, 21-26.

Dancey, C., & Reidy, J. (2017). Hypothesis testing and statistical significance . In Statistics Without Maths for Psychology (7th ed.) (pp. 134-173). Harlow, UK: Pearson.

Hutson, J. (2013). Exploring the nature of lexical selection in spoken word production : competition and the response exclusion hypothesis. England : university of Bristol .

IPIP. (2018). History of the internatinal personality item pool. Retrieved from International Personality Item Pool :

ITK. (2018). About canadian inuit . Retrieved from Inuit Tapiriit Kanatami:

IWH. (2015). Cross-sectional vs longitudinal research . Retrieved from Institute for Work and Health Canada :

Jourdy, R., & Petot, J.-M. (2017). Relationships between personality traits and depression in the light of the “Big Five” and their different facets. Elsevier Masson SAS, 27 - 37 .

Minitab. (2017). What are type I and type II errors? . Retrieved from Minitab:

SD. (2019). Western culture . Retrieved from Science Daily :

Statistics Solutions . (2019). Conduct and interpret factor analysis . Retrieved from Statistics Solutions :

Twenge, J., & Campbell, K. (2017). Personality accross the lifespan . In Personality Psychology . New York : Pearson .

Zhao, X., Solano-Flores, G., & Qian, M. (2018). International test comparisons: reviewing translation error in different source language-target language combinations. nternational Multilingual Research Journal, 17-27.


No comments:

Post a Comment