Descriptive statistics

[Fallon Kathleen Lambert]

Dear Stata Users

What is the best way to report your descriptive statistics using Stata in your report

How do you start writing?

Which variables do you use? All variables that you use in your research?
What about variables you are not sure of using in the end.

How do you incorporate the missings using Stata in your report?

[Attaullah Shah]

For exporting descriptive statistics easily to a Word file, you can use asdoc, that can be installed by

ssc install asdoc

Here are also a few videos on how to quickly get started with asdoc.
https://www.youtube.com/watch?v=zdI65G6AhdU
https://www.youtube.com/watch?v=61ks3cMPz3c

Concerning the question "which variables do you use" it is more related to your research design and the hypotheses, you are testing. Of course, you should report descriptive statistics of variables that help you in answering your research questions.

[Prerak Shah]

Hi Sir,

I have a problem while exporting using “asdoc table x y, c(mean z) row format(%12.2fc) save(results)”. When I open the exported documents, values are rounded off and not the exact ones that we can see on the output window. Could you please help me?

[Sergiy Radyakin]

Isn't it what you added the option format() for??

[Prerak Shah]

Yes, I did, still, it shows the same. Sergiy Radyakin. The problem is not with decimal points, but with the values. So if the value is 412864.32 then it is rounding off to 413000. What should I do?

[Sergiy Radyakin]

Start by compiling a reproducible example using some publicly available data.

[Marcos Almeida]

What is the best way to report your descriptive statistics using Stata in your report

As a starting point, mean + SD for numerical variables and absolute + relative frequency for categorical variables.

Which variables do you use? All variables that you use in your research?
What about variables you are not sure of using in the end.

Think about the Table 1 as a description of the sample. The variables must "make sense" in terms of selection. A glimpse on the literature in the specific field is a good starting point.
The very fact that one collects "n" variables doesn't mean one must use all of them.

How do you incorporate the missings using Stata in your report?

a) You can describe the rate of missing values overall, b) or you can present separated results according to each variable (such as using - codebook - or, better yet, tabulate varname, missing), c) the command misstable summarize will present a detailed description

[Prerak Shah]

I have just put some random values for an example.

3453435.343	1	1	4
5343543.340	1	1	65
3534543.443	2	1	4
3534353.343	2	1	6
3453535.356	3	2	6
56656.650	3	2	7
465446.460	4	2	6
4567566.500	4	2	4

When we run

asdoc table var3 var2[fw=var4], c(mean var1) save(results)

Column1- var1; Column2-var2; Column3-var3; Column4-var4
The ideal result should be same as it will be shown in the output window. However, exported document has some other values. I have checked both in my system. Could you please help me now? Sergiy Radyakin

[Sergiy Radyakin]

I confirm that the output to the Word file does not correspond to the data shown in the output window for the data and command shown above.
-asdoc- contains a significant portion compiled in Mata which only the author can debug.
Please contact the author of this package with a bug report.

Best, Sergiy

[Prerak Shah]

Attaullah Shah Could you please help in here?

[Attaullah Shah]

Please update asdoc my site and comes back with your feedback. If the issue persists, then please use some dataset and post that here with asdoc code to reproduce the error.
The new version of asdoc can be installed from my site. Copy and paste the following line in Stata and press enter.

net install asdoc, from(http://fintechprofessor.com) replace

Please note that the above line has to be copied in full. After installation of the new version, then restart Stata.

Please also visit the new asdocx site here.
https://fintechprofessor.com/asdocx/

[Prerak Shah]

Thank you so much Attaullah Shah. This is working well right now. Thanks for making asdoc and helping student community.