Support #2076
Issues with xx_hadcvvac variables in COVID-19 data collection
50%
Description
Good evening,
I am currently analysing data from the xx_indresp_w datasets of the COVID-19 data collection, specifically from wave 9 (ci), wave 8 (ch) and wave 7 (cg). From the documentation, the questions xx_hadcvvac (about having received the COVID-19 vaccine in each survey wave) should be asked to respondents that have not already answered that they received 1 or 2 doses of vaccines in previous months (answer codes 1 and 2). However, by cross-tabulating the answers to the xx_hadcvvac questions for wave 7 and 9 for respondents present in wave 9 and 7 (left-joining the datasets by respondent ID pidp, i.e. matching all respondents in wave 9 with those that were also in wave 7):
table(ci_hadcvvac = wave_9$ci_hadcvvac, cg_hadcvvac = wave_9$cg_hadcvvac)
with wave_9 the left-joined dataset, I obtain the following table:
cg_hadcvvac
ci_hadcvvac -9 -8 -2 1 2 3 4
-8 0 10 0 133 9 492 4835
-2 2 0 2 0 0 0 4
1 0 0 0 4 1 1 133
2 0 3 1 1663 116 36 2538
3 0 0 0 0 0 1 5
4 0 0 0 2 0 3 322As you can see from the numbers in bold (took as examples), there are some respondents vaccinated in wave 7 that appear to be asked the question again in wave 9. Am I missing some information?
Thank you very much in advance for the support.
Best regards,
Laura
Updated by Understanding Society User Support Team about 1 month ago
- Status changed from New to In Progress
- % Done changed from 0 to 10
- Private changed from Yes to No
Many thanks for your enquiry. The Understanding Society team is looking into it and we will get back to you as soon as we can. We aim to respond to simple queries within 48 hours and more complex issues within 7 working days.
Best wishes,
Understanding Society User Support Team
Updated by Understanding Society User Support Team about 1 month ago
- Status changed from In Progress to Feedback
- % Done changed from 10 to 50
Hello Laura
The question routing is as follows: If ff_hadcvvac = 3, then ask if the respondent has not reported in a previous month that they had a coronavirus vaccine.
When tabulating (tab ci_hadcvvac ci_ff_hadcvvac), we see responses only when ci_ff_hadcvvac = 3, which is expected.
| Had covid-19 vaccine | 1 Yes, first vaccine only | 2 Yes, both vaccinations | 3 No | Total |
|---|---|---|---|---|
| -8 inapplicable | 5,917 | 175 | 24 | 6,116 |
| -2 refusal | 0 | 0 | 10 | 10 |
| 1 Yes, first vaccination only | 0 | 0 | 201 | 201 |
| 2 Yes, both vaccinations | 0 | 0 | 5,972 | 5,972 |
| 3 No, but I have an appointment | 0 | 0 | 13 | 13 |
| 4 No | 0 | 0 | 506 | 506 |
| Total | 5,917 | 175 | 6,726 | 12,818 |
These observations might be related to booster doses, which began in September 2021, six months after their second vaccine dose, to members of the following groups: people over the age of 50, vulnerable people over 16, health and social care workers, adult members of the households of immune-suppressed individuals
If you check the age of the 1,663 (cg_hadcvvac=1 & ci_hadcvvac==2 ) and 116 (cg_hadcvvac=1 & ci_hadcvvac==1) observations, we find that over 80% are above 50 years old, which were the first group receiving the booster does.
I hope this information is helpful.
Best wishes,
Roberto Cavazos
Understanding Society User Support Team
Updated by Laura L about 1 month ago
Understanding Society User Support Team wrote in #note-2:
Hello Laura
The question routing is as follows: If ff_hadcvvac = 3, then ask if the respondent has not reported in a previous month that they had a coronavirus vaccine.
When tabulating (tab ci_hadcvvac ci_ff_hadcvvac), we see responses only when ci_ff_hadcvvac = 3, which is expected.
Had covid-19 vaccine 1 Yes, first vaccine only 2 Yes, both vaccinations 3 No Total -8 inapplicable 5,917 175 24 6,116 -2 refusal 0 0 10 10 1 Yes, first vaccination only 0 0 201 201 2 Yes, both vaccinations 0 0 5,972 5,972 3 No, but I have an appointment 0 0 13 13 4 No 0 0 506 506 Total 5,917 175 6,726 12,818 These observations might be related to booster doses, which began in September 2021, six months after their second vaccine dose, to members of the following groups: people over the age of 50, vulnerable people over 16, health and social care workers, adult members of the households of immune-suppressed individuals
If you check the age of the 1,663 (cg_hadcvvac=1 & ci_hadcvvac==2 ) and 116 (cg_hadcvvac=1 & ci_hadcvvac==1) observations, we find that over 80% are above 50 years old, which were the first group receiving the booster does.
I hope this information is helpful.
Best wishes,
Roberto Cavazos
Understanding Society User Support Team
Dear Roberto,
Thank you very much for your kind answer.
I hope I am not bothering you, but this is still not fully clear to me. Specifically, how are the routing variables ff_hadcvvac constructed?
For instance, from what I understand ch_ff_hadcvvac should be derived by respondents' answer to cg_hadcvvac (except of course for new respondents). However, by tabulating table(cg_hadcvvac = check$cg_hadcvvac, ch_ff_hadcvvac = check$ch_ff_hadcvvac) for respondents in wave 7 (check is the dataset containing the wave 7 respondents), this is what I get:
ch_ff_hadcvvac
cg_hadcvvac -8 1 2 3
-9 0 0 0 2
-8 1 0 0 15
-2 0 0 0 5
1 0 2078 0 31
2 0 0 145 3
3 0 0 0 604
4 9 0 0 9075
How are some respondents assigned to category 3 if they answered that they got the vaccine in wave 7? Also, what does the -8 value mean in the context of routing variables ff_hadcvvac? Lastly, you were telling me that the observations for which cg_hadcvvac=1 & ci_hadcvvac==2 and cg_hadcvvac=1 & ci_hadcvvac==1 might refer to booster doses. However, the question options are:
Have you had a coronavirus vaccination?
1. Yes, first vaccination only
2. Yes, both vaccinations
3. No, but I have an appointment
4. NoHow can they refer to boosters given the wording?
Thank you again for your support and prompt response.
Best regards,
Laura
Updated by Understanding Society User Support Team about 1 month ago
Hello Laura,
Typically, upon receiving information from the field agency, we process the fed forward variables and return them to the fieldwork to implement in the subsequent wave questionnaire. You can verify the coding for variable ff_hadcvvac in the "variables used for routing" section of the wave 8 or 9 questionnaires.
The universe for the question "Had covid-19 vaccine" (hadcvvac) is defined as if ff_hadcvvac = 3, indicating that the respondent has not reported in a previous month that they had received a coronavirus vaccine. For waves 8 and 9, the routing is correct; only observations coded as ch_ff_hadcvvac = 3 have information on ch_hadcvvac, and similarly for ci_ff_hadcvvac = 3 and ci_hadcvvac. However, the routing does not cross-reference the fed forward variable against the actual response in the previous wave, meaning the fed forward variable in wave 9 is compared with the actual response in wave 8. I’m going to ask our Covid-19 team and give you an update as soon as possible.
Regarding the booster dose, what I meant was that respondents who did not receive a vaccine in the previous month of wave 9 (ci_ff_hadcvvac = 3) are reported as having one or two vaccine doses in wave 7 and two vaccinations in wave 9. Therefore, those who received more than two doses are those who received a booster jab. Since we do not have this as a response option, respondents had to choose options 1 or 2. Additionally, around 80% of these respondents were above 50 years old, the first group to receive the booster jab.
Once I get a response from our Covid-19 team I’ll let you know. Please feel free to send us any issues or questions about the Study. Happy to help.
I hope this information is helpful.
Best wishes,
Roberto Cavazos
Understanding Society User Support Team
Updated by Laura L about 1 month ago
Understanding Society User Support Team wrote in #note-4:
Hello Laura,
Typically, upon receiving information from the field agency, we process the fed forward variables and return them to the fieldwork to implement in the subsequent wave questionnaire. You can verify the coding for variable ff_hadcvvac in the "variables used for routing" section of the wave 8 or 9 questionnaires.
The universe for the question "Had covid-19 vaccine" (hadcvvac) is defined as if ff_hadcvvac = 3, indicating that the respondent has not reported in a previous month that they had received a coronavirus vaccine. For waves 8 and 9, the routing is correct; only observations coded as ch_ff_hadcvvac = 3 have information on ch_hadcvvac, and similarly for ci_ff_hadcvvac = 3 and ci_hadcvvac. However, the routing does not cross-reference the fed forward variable against the actual response in the previous wave, meaning the fed forward variable in wave 9 is compared with the actual response in wave 8. I’m going to ask our Covid-19 team and give you an update as soon as possible.
Regarding the booster dose, what I meant was that respondents who did not receive a vaccine in the previous month of wave 9 (ci_ff_hadcvvac = 3) are reported as having one or two vaccine doses in wave 7 and two vaccinations in wave 9. Therefore, those who received more than two doses are those who received a booster jab. Since we do not have this as a response option, respondents had to choose options 1 or 2. Additionally, around 80% of these respondents were above 50 years old, the first group to receive the booster jab.
Once I get a response from our Covid-19 team I’ll let you know. Please feel free to send us any issues or questions about the Study. Happy to help.
I hope this information is helpful.
Best wishes,
Roberto Cavazos
Understanding Society User Support Team
Dear Roberto,
Thank you for your kind and exhaustive answer.
Everything is clearer now. I'll just wait for your update on how the fed-forward variables relate to the variables they should reference from previous waves.
All the best,
Laura