Don't let age fool you

What are the Big 5 Personality traits? 

The Big 5 personalities are "five broad domains that define human personality and account for individual differences (Dr. Edwin van Thiel)."

The dimensions of the personalities may be broad but is supported by years of empirical and data driven research. It was started by the U.S Air Force and is now often used by HR professional to assess whether or not an employee will be fit for their job. 

When an HR professional wants to hire a new employee, some things they may take into account are age and race. Some questions he/she may ask are is this person too young? too old? will his/her personality fit into our company? This person is not from the USA, will his/her personality fit into our company?

The Data

The data was collected from openpsychometrics.org

A website open to the public to educate them 'about various personality tests, their uses and meaning, the various theories of personality.

The data was cleaned and the answers to the tests were calculated using the formula provided by openpsychometerics here.

Visualizing Personality Data

Interact with the graph below and scroll through the pages to see the difference in personalities between the USA and the rest of the world.

Does age affect personality?

Spearman's Correlation Coefficient was used to check the relation between age as a continuous variable and different personality traits. 

According to the heat map above, age has a weak positive relation with all 5 personality types.

Because of the weak correlation I decided to group age into different groups:

minor(<18), ya(18-24), adult(24-40), older adult(40+)

The Kruskal-Wallis test was used to check if the population median of all age groups were equal. A non-parametric test was used because the data did not follow a normal distribution.

age_extroversion : (statistic=0.5302271494403866, pvalue=0.9122021872216677)
age_agreeableness: (statistic=0.7033022906303688, pvalue=0.8724269434161755)
age_conscientiousness: (statistic=2.333199277065723, pvalue=0.5061906590878511)
age_neuroticism: (statistic=0.6542088781149311, pvalue=0.883917873685482)
age_open2experience: (statistic=0.8017139375186264, pvalue=0.8490570383457074)

According to the test results with an alpha of .05, the age groups and all personality traits 

failed to reject the null hypothesis, meaning that the population median of all groups are 

equal.

Age-group does not have a significant impact on the personality scores.

The full code and thought process can be found in my github.

Do personalities differ between countries?

Specifically, USA vs the rest of the world, is the difference between the USA and the rest 
statistically significant?

Mann-Whitney U Test was used to check if the distribution between the two groups were 
equal.

usa_extroversion: (statistic=44753597.5, pvalue=8.895596261785486e-05)
usa_agreeablness: (statistic=40869091.0, pvalue=9.169508717821433e-44)
usa_concientiousness: (statistic=41438557.5, pvalue=2.6100889117150396e-35)
usa_neuroticism: (statistic=42783714.5, pvalue=4.196078925028747e-19)
usa_open2experience: (statistic=43012833.0, pvalue=6.944228123169958e-17)

According to the test results with an alpha of .05, the null hypothesis was rejected. 

The personalities of people in the USA are different than the rest of the world. 

The full code and thought process can be found in my github.


What does this mean?

Age has a weak correlation between personality and there is no statistically significance

between the different age groups and personality.

However, there is a statistically significant difference between the participants personalities 

of the USA vs the World.

When looking into new employees, an HR professional should take into account the 
country from where the candidate is from. Are they from the US or not?

Limitations

Some limitations with this experiment was that the data was collected from a USA based site, which means that most participants came from the USA. We can not compare each country 

because we have limited data. Another limitation is the age group. The age groups were superficially made and in future experiments we can increase or decrease the number of groups, 

and/or change bin lengths for each age.

Future

To get more meaning from the results of the statistical analysis we can dive deeper into each 

personality trait and their respective score values. With more data we can also look into the 

personality differences between every country instead of just two.

Source

https://www.123test.com/big-five-personality-theory/

https://openpsychometrics.org/printable/big-five-personality-test.pdf

https://github.com/yoonsunghwa