Project

General

Profile

Actions

Support #2231

open

Partner's information (indresp or indall)

Added by Luis Ortiz 29 days ago. Updated 14 days ago.

Status:
Resolved
Priority:
Normal
Category:
Data management
Start date:
03/26/2025
% Done:

100%


Description

Dear colleagues,

For research I'm doing with UKHLS, I'm interested in matching partners' information on gender egalitarian values (scopfamb & scopfamf) and education (qfhigh_dv) to ego's information. For instance, I want to get information about the male partner's gender egalitarian values and level of education, but also about the female gender egalitarian values and level of education, if the ego is male. For this purpose, I'm using the wonderful guidelines provided by your team here: https://moodlex.essex.ac.uk/course/view.php?id=76

The guidelines are very useful, but I've got doubts regarding the convenience of using 'indresp' or 'indall' as the source of information where to draw information about the partner for every wave. If I'm not wrong, in 'indresp', partner information is only available if the respondent reported having a co-resident partner and that partner is enumerated in the household. 'indall' includes everyone listed on the household grid, even if they were not eligible or didn't respond to the individual interview.

I am aware that some of the information I'm interested is not available in all the waves (scopfamb & scopfamf). For that reason I wouldn't like to lose even more information because I'm not able to identify the partner because I'm not using the proper data files to retrieve their information from.

Thanks a lot for your attention

And kind regards

Luis Ortiz
Universitat Pompeu Fabra


Files

Compile individual info.do (3.87 KB) Compile individual info.do Luis Ortiz, 04/09/2025 05:05 PM
Compile partners info.do (6.92 KB) Compile partners info.do Luis Ortiz, 04/09/2025 05:05 PM
Variables and table.do (3.65 KB) Variables and table.do Luis Ortiz, 04/09/2025 05:05 PM
Actions #1

Updated by Understanding Society User Support Team 29 days ago

  • Category set to Data management
  • Status changed from New to Feedback
  • % Done changed from 0 to 80
  • Private changed from Yes to No

Dear Luis,

All these variables are only in indresp, so you will need to use it for your analysis. While indall is more complete in terms of the people included, it contains only basic socio-demographic information. Also, I’d recommend checking whether the hiqual_dv variable could be used in your project, as it is the standard education variable we usually recommend using.

Best wishes,
Piotr Marzec
UKHLS User Support

Actions #2

Updated by Luis Ortiz 28 days ago

Many thanks, Piotr...¡

Your answer is very helpful.

Kind regards

Luis Ortiz

Actions #3

Updated by Understanding Society User Support Team 23 days ago

Hi Luis,

You're welcome!

Best wishes,
Piotr

Updated by Luis Ortiz 15 days ago

Dear Piotr (and colleagues),

I greatly appreciate your support in my previous query.

As I said then, 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)


| RECODE of _sp_highest_educ | No partner Less than HE Higher educ Total
----------------+-------------------------------------------------
gndr_educ_group |
Men (Mid) | 4.2 5.3 1.5 11.0
Men (High) | 4.4 5.0 7.2 16.5
Women (Mid) | 17.1 14.1 4.7 35.9
Women (High) | 12.6 9.7 14.3 36.6
Total | 38.3 34.0 27.7 100.0
------------------------------------------------------------------

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

I have summarised the do-files that I've created for compiling individual-level information of the 14 waves of the UKHLS, retrieving and merging partner's information (second do-file) and, finally, creating the variables that participate in the creation of the table above, which is the last line in the third do-file.

My co-authors and I would greatly appreciate it if you helped us to solve this puzzle.

Thanks a lot for your attention and your help, which is invaluable.

And kind regards

Luis Ortiz
Universitat Pompeu Fabra

Actions #5

Updated by Luis Ortiz 15 days ago

Luis Ortiz wrote in #note-4:

Dear Piotr (and colleagues),

I greatly appreciate your support in my previous query.

As I said then, 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)


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

------------------------------------------------------------------
                |             RECODE of _sp_highest_educ          
                |  No partner   Less than HE   Higher educ   Total
----------------+-------------------------------------------------
gndr_educ_group |                                                 
  Men (Mid)     |         4.2            5.3           1.5    11.0
  Men (High)    |         4.4            5.0           7.2    16.5
  Women (Mid)   |        17.1           14.1           4.7    35.9
  Women (High)  |        12.6            9.7          14.3    36.6
  Total         |        38.3           34.0          27.7   100.0
------------------------------------------------------------------

. 

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

I have summarised the do-files that I've created for compiling individual-level information of the 14 waves of the UKHLS, retrieving and merging partner's information (second do-file) and, finally, creating the variables that participate in the creation of the table above, which is the last line in the third do-file.

My co-authors and I would greatly appreciate it if you helped us to solve this puzzle.

Thanks a lot for your attention and your help, which is invaluable.

And kind regards

Luis Ortiz
Universitat Pompeu Fabra

Actions #6

Updated by Understanding Society User Support Team 14 days ago

  • Status changed from Feedback to Resolved
  • % Done changed from 80 to 100

Hi Luis,

As this is continuing separately as Issue 2236, I setting this to resolved.

Best wishes,
Alita

Actions

Also available in: Atom PDF