Introduction
The article called “Assessing the Big Five Personality Traits with Latent Semantic Analysis” by Kwantes, Derbentseva, Lam, Vartanian, and Marmurek (2016) explores whether or not the individuals’ personality features can be evaluated based on the analysis of content and semantics of their written text. The study targets the possibility of discovering one’s personality based on how they write and which words they choose. The research question and focus of the article are interesting and important because the body of literature on this particular subject is insufficient. The authors collect the data and present it in a clear manner and make a valuable point; however, the research has definite limitations due to such aspects as the subjective interpretation of the results, small number of participants, and unclear transferability. Overall, the research touches on an important topic but the techniques to study the results and to adjust the assessment still need improvement.
Description of the Paper Selected
Big Five personality traits have been found to be in correlation with various types of performance – professional, academic, lifestyle choices, social interactions, and even political preferences and voting behaviors, to name a few (Zhou & Schwenker, 2013; Gerber, Huber, Doherty, Dowling, & Panagopoulos, 2011; Shaw, Wu, Irwin, & Patrizi, 2016). That is why the ways to measure these personality domains have been treated and researched as an issue of high importance; at the same time, the multitude of studies have brought fragmentation to this sphere of knowledge instead of creating universally applicable approaches and assessments (John & Srivastava, 1999). Kwantes et al. (2016) investigate whether or not it is possible to assess the five personality domains by means of analyzing the tests written by people. Potentially, this approach could serve as a universal type of assessment that can be used in versatile spheres. The current Big Five Inventory is a lengthy assessment that contains over forty items that usually is carried out in a form of a self-report (John, Naumann, & Soto, 2008). The authors offer a new solution that is much more flexible – the assessment based on texts written by the participants. For that purpose, the researchers selected 115 participants (undergraduate students, ages 18 to 23) and first assess their Big Five personality traits using the standard Big Five Inventory with 44 units mentioned previously. Over the course of the research, the number of students who completed all the stages of assessment went down to 87. The participants were asked to respond to five different scenarios in a form of a written text; the scenarios are designed to correlate with the Big Five personality traits (one scenario per each trait) and place the participants in the situations as active agents (Kwantes et al., 2016). The finished written responses of the participants were then evaluated using the latent semantic analysis (or LSA) for a purpose to detect the use of specific words and distinguish between the word choice and the Big Five personality traits.
Evaluation and Argument
The Big Five personality traits are Extraversion (sometimes called urgency). The broad dimension of Extraversion (energy, communicability), agreeableness (affection and kindness), conscientiousness (planning and organizational abilities), neuroticism (emotional responses, moods, anxiety), and openness (willingness to have new experiences) (Srivastava, n. d.; Goldberg & Rosolack, 1994).
Rationale
The rationale behind the research presented by Kwantes et al. (2016) is based on the need for a faster and more flexible way to assess the Big Five personality traits of the individuals. This need has been identified by Gosling, Rentfrow, and Swann (2003), who specified that the assessors have to choose whether the domains should be evaluated using the lengthy test or not evaluated at all. The information received due to the assessment is very important because it allows the predictions of behavioral, affective, and cognitive patterns of the participants (Zillig, Hemenover, & Dienstbier, 2002). The predictive power of the current inventory is quite high, but its disadvantage is the length and the form of a self-report (Soto & John, 2016). That is why a new type of assessment is needed.
Sources
In the introduction section, Kwantes et al. (2016) present the findings of the previous studies conducted and published throughout the 1990s and the beginning of the 2000s. In particular, they explore the language-based assessments and the correlation between their results and the actual behaviors and cognitive features of the evaluated individuals. The researchers explore the use of latent semantic analysis as a reliable way to analyze written texts and make conclusions as to the personality traits of the authors. All in all, Kwantes et al. refer to many scholarly works dedicated to the subjects relevant to the research question and objective and establish that the use of the latent semantic analysis for a purpose to measure the Big Five personality domains is majorly unresearched and carries the potential. That is why the authors make a conclusion that a thorough exploration of this subject could bring valuable results and open a new field for future studies. The sources referenced by Kwantes et al. are mostly recent; however there is one source from the 1980s, and two sources from the 1930s and 40s that can be considered irrelevant due to their dates. At the same time, these sources were used to find information about the personality trait theories in psychology and thus can be viewed as valuable and reliable.
Methodology
The researchers relied on qualitative methods of data collection and analysis. The researchers relied on the BFI to assess the domains of the participants and then evaluated their written texts by means of finding the words and the contents that would correlate with the overall vector of the domain. That way, no specific coding of words was used, instead, the authors assessed the quality of the words in the texts based on the intuitive principles. Kwantes et al. (2016) indicated that their results were consistent with those of similar research conducted earlier.
Participants
The participants of the study were Psychology students. Their number was relatively small (87 people). Also, all of them reported that English was their first language. That way, the sample was chosen using convenience criteria because assessing the written responses to the personality domain scenarios provided by the non-native English speakers could have been confusing due to the language barriers and limitations.
Results and Implications
The findings showed that the scenarios targeting the domains of Agreeableness and Conscientiousness were not in correlation with the BFI results of the participants. The implication is that the scenarios were not consistent with the assessed traits and require adjustments. Theoretically, the approach sounds like a fast and relatively simple way to assess the domains of personality. However, practically, it is difficult to interpret and is likely to generate vague and unclear results that would not correlate with reality.
Limitations
There were several factors working as the limitations to the study. First of all, it is possible to fake the assessment based on the LSA; and since the participants were psychology students, it was possible for them to alter their responses in a way that would distort the results (Ziegler, MacCann, & Roberts, 2012).
Secondly, the semantic analysis of the written responses to the scenarios was conducted based on the researchers’ intuitive perceptions of the words used by the participants. This approach assumes that biased evaluation was possible and the interpretation of the results could have been incorrect.
Thirdly, the research only included individuals whose first language is English, which means that the research may not be applicable to diverse groups of participants.
Finally, two out of five personality domains were not in correlation in the written texts and the inventory results. That is why the research techniques or materials can be evaluated as partly ineffective.
Conclusion
To sum up, Kwantes et al. (2016) presented substantial research in the field that has not been explored yet, thus, encouraging future studies and the possible development of another inventory for the Big Five personality traits assessment. Currently, the methods and techniques they offered proved to be partially unsuitable and in need of adjustments in order to improve the reliability and validity of the assessment results. In addition, different types of limitations and biases were possible over the course of the presented research. They include the sample size and the participants speaking English as their first language, the subjective interpretation of the texts by the researchers, the students’ ability to trick or fake the inventory responses and distort the results, and the inconsistency between texts and BFI results in two out of five domains.
References
Gerber, A., Huber, G., Doherty, D., Dowling, C., & Panagopoulos, C. (2011). Big Five Personality Traits and Responses to Persuasive Appeals: Results from Voter Turnout Experiments. Web.
Goldberg, L. R. & Rosolack, T. K. (1994). The Big Five factor structure as an integrative framework. Web.
Gosling, S., Rentfrow, P., & Swann, W. (2003). A very brief measure of the Big-Five personality domains. Journal of Research in Personality, 37(6), 504-528.
John, O. P., Naumann, L. P., & Soto, C. J. (2008). Paradigm Shift to the Integrative Big-Five Trait Taxonomy: History, Measurement, and Conceptual Issues. In O. P. John, R. W. Robins, & L. A. Pervin (Eds.), Handbook of personality: Theory and research (pp. 114-158). New York, NY: Guilford Press.
John, O. P., & Srivastava, S. (1999). The Big-Five Trait Taxonomy: History, Measurement, and Theoretical Perspectives. Web.
Kwantes, P., Derbentseva, N., Lam, Q., Vartanian, O., & Marmurek, H. (2016). Assessing the Big Five personality traits with latent semantic analysis. Personality and Individual Differences, 102, 229–233
Shaw, C., Wu, X., Irwin, K., & Patrizi,L. (2016). Faculty Personality: A Factor of Student Retention. Online Journal of Distance Learning Administration, 19(1), n. p.
Soto, C. & John, O. (2016). The Next Big Five Inventory (BFI-2): Developing and Assessing a Hierarchical Model With 15 Facets to Enhance Bandwidth, Fidelity, and Predictive Power. Journal of Personality and Social Psychology. Web.
Srivastava, S. (n. d.). Measuring the Big Five Personality Domains. Web.
Zhou, Z. & Schwenker, F. (2013). Partially supervised learning. New York, NY: Springer.
Ziegler, M., MacCann, C., & Roberts, R. (2012). New Perspectives on Faking in Personality Assessment. New York, NY: Oxford University Press.
Zillig, L., Hemenover, S., & Dienstbier, R. (2002). What Do We Assess when We Assess a Big 5 Trait? A Content Analysis of the Affective, Behavioral, and Cognitive Processes Represented in Big 5 Personality Inventories. Personality and Social Psychology Bulletin, 28(6), 847-858.