The Finishing Touches

Mahita v -

Welcome back to week 8 of my senior blog posts. This week I finished up my analysis for both of my survey themes. I also went back and conducted chi-square tests to clarify the results of my logistic regression. I used a Chi-Square Test of Independence to examine whether a patient’s race was associated with each of the insurance-related variables in breast cancer care. Since both variables were categorical, this test helped me determine if the disparities I observed were statistically significant — not just random chance.

I researched what would be the best way to conduct the tests, and discovered that I could use the scipy and researchpy libraries in python. 

—————————————————————————————————————–

crosstab = pd.crosstab(df_insurance[“race”], df_insurance[“insurance_acceptance”])

Crosstab

stats.chi2_contingency(crosstab)

crosstab, test_results, expected = rp.crosstab(df_insurance[“race”], df_insurance[“insurance_acceptance”],

                                               test= “chi-square”,

                                               expected_freqs= True,

                                               prop= “cell”)

crosstab

test_results

——————————————————————————————————————

I found that there was a statistically significant difference for all the questions except insurance acceptance (which matches the results from the logistic regression). The code also gave me a value for Cramer’s V –  a statistical measure that tells us the strength of the association between two categorical variables ranging from 0 to 1. For instance, the Cramer’s V for the copay question was 0.0537 which suggests a minimal influence of race and ability to afford the copay. This statistical measure helps me understand the extent of the statistical significance. However, I’m planning to have a meeting with my project advisor to look over the results as well.

Other than that, I also finished creating my data tables and starting my slides for my presentation. Next week, I will be fully focused on writing my final research paper and practicing my presentation.

More Posts

Comments:

All viewpoints are welcome but profane, threatening, disrespectful, or harassing comments will not be tolerated and are subject to moderation up to, and including, full deletion.

    christina_v
    Hi Mahita! Love how you used both logistic regression and chi-square tests to strengthen your analysis. Cramer’s V is such a cool addition too. Were you surprised that insurance acceptance wasn’t significant? Can’t wait to see how your final paper and presentation turn out!
    mahita_v
    Yes, I was a little surprised! I think this trend might either be due to a white-dominant database as well as issues with insurance acceptance across the healthcare system - highlighting an even broader problem.

Leave a Reply

Your email address will not be published. Required fields are marked *