[Sherry Ahmed]

Writing my results now and it came to my mind that. Can someone argue that these results are not important because, for example, compared to a male, the probability a female to commute above 100 km is lower by 0.0718827 percentage points. Can the argue that this number does not represent anything important? is there a way to report these number as a percent instead of a percentage points for audience to understand it? I'm just worried that they can argue that (0.07 is a very small number to even mention... which is my case in all the variables)?

[Sherry Ahmed]

Just to make sure I'm writing this right.... compared to a male, the probability a female to commute above 100 km is lower by 0.0718827 (which means lower by 7 percentage points NOT 0.07 percentage point). Is that correct? but still, this percentage point (even after multiplying my marginal effects results x 100) will stay very small. Having kids as an example will be (-0.002 which means 0.2 percentage points). Can it be argued that they are really meaningful?

[Clyde Schechter]

First, yes, -.0718827 is 7 percentage points, not .07 percentage points.

Importance, or meaningfulness, is a judgment call that people familiar with the context have to make. And reasonable people can disagree in those judgments. You can have a discussion with people about how meaningful a 7 percentage point difference is, but it won't be a statistical discussion--it will be a discussion about the utility and disutility of commuting more than 100km, and about the costs and benefits of getting married, having kids, obtaining a bachelors degree and so on, are.

[Sherry Ahmed]

Perfect. Thank you so much

[Sherry Ahmed]

Sorry to bother you again but I Just have a quick question regarding a different topic: I created a different post because I'm confused about another regression (Blinder Oaxaca wage decomposition) but I did not get any reply. Was that wrong (to create a new topic)? Should I have to only have one thread and add my questions here?

[Clyde Schechter]

You were absolutely correct to create a new thread for a new topic. It's very important to keep threads on topic. People come to the Forum to search for specific information, and if the threads digress, then they will either fail to find what they search for, or waste time reading things that are irrelevant to their search. So you did the right thing.

I'm not going to respond to that thread because I know zero about Blinder Oaxaca wage decomposition. But there are others who know. Be patient. If your question is clearly posed, you'll get a response. If you don't get one within a reasonable time (say 24-36 hours), don't just bump your post. Revise it to be clear and include all needed information. If you aren't sure about what to say, try reading the Forum FAQs for advice.

[Sherry Ahmed]

Thank you so much. I included as much information and data as I can as well as my regression results. I will keep monitoring it and hopefully I will get a response soon.