Project

General

Profile

Support #132

problem obtaining the numbers for each ethnic group in in wave 2

Added by Stef Doebler almost 7 years ago. Updated almost 7 years ago.

Status:
Closed
Priority:
Normal
Assignee:
Redmine Admin
Category:
Ethnicity
Target version:
Start date:
03/14/2013
Due date:
% Done:

100%

Estimated time:

Description

Dear Understanding Society Team,Dear All,
I am trying to with the Understanding Society data wave 2. I am mainly interested in differences between ethnic groups.

I have a problem when using wave 2 and I am wondering if I am doing something wrong. The data documentation (user guide etc) did not help me at all with this, nor does emailing . The only response I received was to post my question here, and the response was not signed by a person. I'd be really grateful for any help!

I am interested in the individual level, not the household at this point. So I am using the b-indresp data file. I am trying to use the ethnic booster sample in wave 2. When requesting the frequencies of the household origin variable, the the output tells me that 5598 respondents are members of the ethnic booster sample.
However, when I then request the frequencies for the ethnic group indicator variable b_racel , then I get the following: 46811 respondents are in a category called ' inapplicable' and 3882 respondents are in a category called 'proxy'. However, the numbers of respondents belonging to each ethnic category of that variable (b_racel) are far too small, they are nowhere near the 1000+ respondents that (according to the userguide) were sampled for each ethnic group.
My questions are thus:
1. what does the category ,inapplicable' mean? Who is in that category and why is this category part of the ethnic minority variable? (this category is not part of the ethnic minority variable in wave 1). Unfortunately I did not find the answer in the data documentation.
2. what does the category 'proxy' mean?
3. where is the ethnic booster? How can I get the ethnic booster and what do I need to do to obtain the numbers of respondents belonging to each ethnic group using the 2nd wave of 'Understanding society'? How can I get the numbers of respondents belonging to each ethnic group? Requesting the frequencies of variable b_racel does not work.

I can get the ethnic booster without problems in wave 1 but not in wave 2.

Many thanks in advance for your help.

Best,
Stef

History

#1 Updated by Redmine Admin almost 7 years ago

  • Category set to Ethnicity
  • Status changed from New to In Progress
  • Assignee set to Redmine Admin
  • Target version set to M2

Stef,

1. what does the category ,inapplicable' mean? Who is in that category and why is this category part of the ethnic minority variable? (this category is not part of the ethnic minority variable in wave 1).

The ethnic group question (_racel) is asked during adult personal interviews, but not of respondent interviewed by proxy. The questionnaire design is dependent interviewing, so only respondents who have not been interviewed before are asked the question. The XWAVEDAT file consolidates valid information obtained across waves for racel, which might be of use in this case. Please be aware that ethnic group information is not collected on children.

2. what does the category 'proxy' mean?

A subset of questions may be answered by another adult in the household if the respondent is away.

3. where is the ethnic booster? How can I get the ethnic booster and what do I need to do to obtain the numbers of respondents belonging to each ethnic group using the 2nd wave of 'Understanding society'? How can I get the numbers of respondents belonging to each ethnic group? Requesting the frequencies of variable b_racel does not work.

The ethnic minority boost sample can be identified by the variable, _hhorig. The user guide explains more about the weights supplied with the dataset, which will allow you to study target ethnic groups together with the general population sample.

Jakob

#2 Updated by Redmine Admin almost 7 years ago

  • % Done changed from 0 to 50

#3 Updated by Stef Doebler almost 7 years ago

Dear Jakob, Dear All,

Many thanks for your help. Jakob.

I just have a follow up question: I know that _hhorig (the 'household origin variable') gives me the number of cases that belong to the ethnic booster sample. I have requested the frequencies of this variable and as said above 5598 respondents are in the ethnic bosster. However,when I request the frequencies for b_racel, the variable that according to the userguide is supposed to contain the members of each ethnic minority, then I get the following: 46,811 respondents are in a category called ' inapplicable'and the numbers of respondents belonging to each ethnic category of b_racel are nowhere near the 1000+ respondents that (according to the userguide) were sampled for each ethnic group. For example: b_racel gives out 195 respondents are Indian, 184 respondents are Pakistani, 185 respondents are Bangladeshi. So it is obvious that b_racel does not include the members of the ethnic booster. Thus my question is still: what variable do I need to use and what do I need to do to obtain the numbers of respondents in wave 2 who belong to each ethnic group. B-Racel does not give me this information. (I read the userguide and also the part on weighting and it does not contain the answer). This problem does not occur in wave 1, only in wave 2.
Many thanks for your help!
Best Wishes,
Stef

#4 Updated by Redmine Admin almost 7 years ago

As suggested above, I would look at the variables hhorig and racel on XWAVEDAT (you will see that we are close to the 1000 target). You may also create your own cross-wave dataset combining the racel information given when first interviewed (cf. dependent interviewing design). Do let us know if this does not quite answer your question.
Jakob

#5 Updated by Stef Doebler almost 7 years ago

Many Thanks, Jakob!! This is very helpful. Now I understand.
Best Wishes,
STef

#6 Updated by Redmine Admin almost 7 years ago

  • Status changed from In Progress to Closed
  • % Done changed from 50 to 100

Also available in: Atom PDF