Based on your description, it sounds like you have data on the number of subscribers and non-subscribers before and after a stimulus (time pressure or method) was applied, and you want to determine which stimulus has the greatest effect on the number of subscribers and non-subscribers.

One statistical technique you could use to analyze these data is a 2-way ANOVA (analysis of variance) with interaction. This would allow you to test for the effect of time pressure (weekly vs. monthly) and method (method 1 vs. method 2) on the number of subscribers and non-subscribers separately.

Additionally, you could also include a factor in your ANOVA design to differentiate between subscribers and non-subscribers. This would enable you to test whether the effect of the stimulus on the number of subscribers is different from the effect of the stimulus on the number of non-subscribers.

You could also use the chi-squared test for independence to determine whether there is a difference in the percentage of decrease in subscriptions between groups of users. This test allows you to compare the proportion of decreases in subscriptions between two groups (subscribers and non-subscribers) for each type of stimulus (week, month, method 1, method 2).

Another option would be to use a multi-level model, that allows you to estimate the effect of time pressure and method on the number of subscribers and non-subscribers. This model would account for the nesting of the observations within the time pressure or method and would allow you to estimate differences between the groups.

It's important to note that any statistical analysis should be carefully planned and executed, and the assumptions of the chosen test should be met. Additionally, is important to consider that, depending on the sample size and the number of observations, other statistical tests may be more appropriate.

Thank you Shezan Ahmed for your answer.

Exactly I want to determine which stimulus has the greatest effect on the number of subscribers and non-subscribers.

I'm unsure if I can use ANOVA because all my variables are categorical, specifically dichotomous.

I will seek information on the chi-square test of independence and the multilevel model to see if they could be used in this research. I would be very grateful if you know some articles or books with examples.

Thank you again.

Diana

For articles or books with examples of these methods, you can search for publications in statistics or methodology journals such as the Journal of Applied Statistics, the Journal of Statistical Computation and Simulation, or the Journal of Educational and Behavioral Statistics. Some articles or books on these topics

• Multilevel Analysis: An Introduction to Basic and Advanced Multilevel Modeling

  Article Multilevel Analysis: An Introduction to Basic and Advanced M...

• "Categorical Data Analysis" by Alan Agresti
• "Chi-Square Goodness of Fit Tests" by Thomas W. O'Gorman

You can also search for these topics on online platforms like JSTOR, SpringerLink, and Google Scholar to find relevant articles.

Hello Diana,

Greetings from NEPAL. You may explore Pre and Post-test statistical techniques to analyze your data as follows: https://eq4peace.org/a-study-of-anger-cognition-in-nepal/

Please feel free to ask any questions, if you have further...

Thanks for your answer Dess Mardan Basnet

I read it fast, so I probably need to read it again slower, but you only compared the percentages obtained, right? Did you use any tests? For example, chi-square?

Shezan Ahmed, I have one question:

Can I use two-way ANOVA when I have a binary (dummy) outcome? I can't find any example of ANOVA with a binary dependent variable.