Attitudes on Aging Study

2024

A survey run in conjunction with McMaster University Professors Len Waverman and Eva Klein.

This project studied UK and Canadian seniors, and measured their attitudes toward aging. The survey size included roughly 4500 participants, but after data cleaning and processing the final sample size was closer to 2000.

Using Scaled Insights’s NLP AI tool, I analyzed the text written by survey respondents, getting personality profiles for each.

I used clustering to segment the data, and a clear pattern emerged. Those who had more pessimistic personality traits, such as neuroticism and coldness, were more likely to think poorly of aging. While those who had higher scores for positive traits had much more optimistic approaches towards ageing.

All findings were summarized through a PowerPoint.

Demographics Surveying:

This survey also included a thorough demographics analysis. This accounted for:

  • Age

  • Gender

  • Disability

  • Pre-existing health condition

  • Location

  • Estimated income

To get the estimated income, I used publicly available census data for both UK and Canada. The Index of Multiple Deprivation estimates how deprived citizens are, and each postcode is given a score on this Index.

By joining the IMD Census data with our respondent’s locations, I had a way to measure and estimate this.

The heatmaps I made for the demographics analysis are interactive, and can be zoomed in on for more detail. This can’t be shared publicly, but I’m always looking for a more dynamic way to visualize data.

This project used a Python data pipeline for NLP-based personality scoring, clustering, demographics analysis, with statistical testing (t-tests, chi-square tests, Cruskal-Wallis ANOVA) to prove significance.

97% of the personality traits were significantly different between clusters, and all measures of attitudes on aging were also significantly different,

Some differences between clusters were insignificant, such as Gender.