enter image description hereQuestion: I am not sure how to describe the sample graph attached. Can you please help me identify the type of plot and how to statistically measure the relationship between the dependent variable (Y-axis) Category A vs Category B?

What success looks like for me: Once I understand how to describe the plot.

  • Is there a statistical method (preference in python) out there that can help me measure the relationship between the two categories of data (Category A & B) .
  • Strength of the relationship between the two categories (Category A & B), positive or negative relationship?
  • List item

My goal with the plot is :

  • Y-axis: One dependent variable (positive real number) example is 'Average Number of days' required to complete one unit of work.

Consideration for :Y-axis: are as below:

  • The dependent variable (y) can however be broken into two or more categories.
  • One example of category is 'Worker type' = Employee or contractor. Employee can be Category A, Contractor can be Category B.
  • Second example of category is 'Country type' = India or USA. India can be Category A, USA can be Category B.
  • As such I can take any category of data from my dataset and split it into various categories (Co-located team v/s distributed teams, Standard team vs Non standard teams) etc.


  • continuous time series data. it can take shape of months, quarters or weeks etc.

Sample plot

  • $\begingroup$ It isn't clear what you are asking here. Are you asking how to determine if the distribution of y is the same for category-A and category-B? In which case, a box-plot would be more helpful than the line plot you've shown. Or are you asking how to determine if the relationship between x and y is different for category-A and category-B, which is what the line plot shows? $\endgroup$
    – Lynn
    Jun 4, 2023 at 3:00

1 Answer 1


Not sure what your end goal is, but if you want one statistical variable to measure how similar the two values are, you can use Null hypothesis and p-value

Null hypothesis - says there is no statistical significance between the two events in the hypothesis.

p-value - which is basically a measure of probability that the null hypotheses is true. A p-value that is less than or equal to 0.05 usually indicates that there is strong evidence against the null hypothesis.

1 - cases are identical

0 - cases are completely different

Read more about using p-value in python.

You can also take a look at Fréchet distance

  • $\begingroup$ Hi Silver Light, Thanks for your feedback. Here's my end goal. Updated my original post with sample dataset. At any given point in time, I would like to compute (using statistical methods like p-value etc) which type of teams are better, more effective, faster, measured by dependent variable (on Y-axis) Lets state a sample Hypothesis - "Are 'ABC' type of software delivery teams more effective in delivering software (measured by 'Y' dependent variable over a time series Quarters, months, weeks etc) $\endgroup$
    – Leo82
    May 24, 2023 at 19:33
  • $\begingroup$ Such that: 'ABC' is any categorical data, in my dataset aka. 'Timezone_synergy' = colocated, overlapping, opposite-timezone etc. plotted on Y axis. Cycle_time is positive real number data, y-axis Quarter/Time - Time series in X-axis $\endgroup$
    – Leo82
    May 24, 2023 at 19:33
  • $\begingroup$ Its almost as if , in doing a side by side / comparative analysis of line graphs aka Catg A & Catg B - instead of manually plotting, viewing them, alternatively do a statistical weighted summary: Such that I am able to notice whether Category A is higher (improving relationship over time) against Category B lower (and weak/worse relationship over time) plotted against a common Y axis dependent variable? Does this make sense? $\endgroup$
    – Leo82
    May 24, 2023 at 20:00

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.