Project

General

Profile

Actions

Support #2236

open

Problems with partner's information

Added by Luis Ortiz 6 days ago. Updated 6 days ago.

Status:
Feedback
Priority:
High
Category:
-
Start date:
04/10/2025
% Done:

50%


Description

Dear colleagues,

[I have previously posted this query as an extension (modification) of a previous one that was solved by Piotr but I thought it advisable to just make a new one out of it. Thanks for your attention]

I'm interested in matching partners' information on gender egalitarian values (scopfamb & scopfamf) and education (qfhigh_dv) to ego's information.

Yet, after retrieving partner's information, I'm getting some unsettling results, like the ones shown in the following table, where four groups created according to gender and level of education of the ego (only middle and highly educated are considered) is cross-tabulated with partner's level of education, considering here three possible categories: no partner, a partner less than HE (higher education) and a partner with higher education.

The results in the following table look as if women had more partners than men; or, putting it differently, as if the observations corresponding to individuals who appear as "no partner", partner with "less than HE" or partner with "Higher Educ" were mostly observations for women (either Mid or High) not for men.

. table gndr_educ_group _sp_highest_educ3 if highest_educ3 > 1, statistic(percent gndr_educ_group) nformat(%5.1f)


As I say, that's unsettling. We would like to know if we made any mistake in retrieving the partner's information or if the partner's data is more available for women than for men.

I have summarised the do-files that I've created for (a) compiling individual-level information of the 14 waves of the UKHLS, (b) retrieving and merging partner's information to the individual-level data created in the previous do-file and (c) creating the variables that participate in the table above. This is the last line in the third do-file.

My co-authors and I would greatly appreciate your help for solving this puzzle.

Thanks a lot for your attention and your invaluable help.

And kind regards

Luis Ortiz
Universitat Pompeu Fabra


Files

clipboard-202504100943-zb70j.png (13.4 KB) clipboard-202504100943-zb70j.png Luis Ortiz, 04/10/2025 08:43 AM
Compile individual info.do (3.87 KB) Compile individual info.do Luis Ortiz, 04/10/2025 08:47 AM
Compile partners info.do (6.92 KB) Compile partners info.do Luis Ortiz, 04/10/2025 08:47 AM
Variables and table.do (3.65 KB) Variables and table.do Luis Ortiz, 04/10/2025 08:47 AM
UserForum_2236_shared.do (1.08 KB) UserForum_2236_shared.do Understanding Society User Support Team, 04/10/2025 09:44 AM
Actions #1

Updated by Understanding Society User Support Team 6 days ago

Hello Luis,

I have attempted to replicate this exercise (Stata syntax attached) for Wave 2. I too am getting higher no partner % for women than men but the discrepancy is not that big - men (2.3%) and women (4.9%). As attrition is higher among separated men, it makes sense that % no partner will be higher for women in the sample.

Please check the syntax I have attached and see if that does what you want to do. If not, please let me know and I will check again.

Hope this helps.
Best wishes,
Alita
Understanding Society User Support Team

Actions #2

Updated by Understanding Society User Support Team 6 days ago

  • Private changed from Yes to No
Actions #3

Updated by Luis Ortiz 6 days ago

Many thanks for your swift attention to our query, Alita...ยก

Your answer is reassuring. The explanation provided for the higher no partner % information for women than for men is very informative. We did not know that attrition among single separated men was higher.

Best wishes and many thanks again

Luis

Actions

Also available in: Atom PDF