Understanding Society User Support: Issueshttps://iserredex.essex.ac.uk/support/https://iserredex.essex.ac.uk/support/support/favicon.ico?15995719382024-02-20T14:23:07ZUnderstanding Society User Support
Redmine Support #2056 (Feedback): Which weights to use when combining the mainstage and Covid-19 waves of...https://iserredex.essex.ac.uk/support/issues/20562024-02-20T14:23:07ZJames Laurence
<p>Hi there,</p>
<p>I have been reading through the forum support on weighting when using the Covid-19 data and Mainstage data combined, but I was hoping to just get some clarification on a couple of points.</p>
<p>I am conducting a study which looks at trends over time in adult mental health across UKHLS Mainstage surveys 2, 3, 4, 5, 6, 7, 8 and 9, and the Covid-19 survey 1, 2, 3, 4, 5, 6, 7, 8 (only the web survey). I end at Mainstage survey wave 9 (survey period Jan 2017-May 2019) as that appears to be the last Mainstage wave that doesn’t cover the pandemic period. My data is in long format and I’m using Stata.</p>
<p>However, I am unsure of which weights to use precisely for my aims.</p>
<p>The first thing I would like to do is to treat the data as pooled cross-sectional data to look at how mental health changes over time (between wave 2 of the mainstage and wave 8 of the Covid survey). So, I’d like to see the level of mental health in the UK at each wave of the UKHLS. Ideally, I would like to model all the waves together. Am I right in thinking that to do so I would need to create a new ‘weight’ variable, which is the self-completion cross-sectional weight for each wave? I’ll explain below…</p>
<p>So, for wave 2 of the Mainstage survey, the new ‘weight’ variable would have the value of b_indscub_xw<br />For wave 3 it would have the value of: c_indscub_xw<br />For wave 4 it would have the value of: d_indscub_xw<br />For wave 5 it would have the value of: e_indscub_xw<br />For wave 6, it seems there are two self-completion, cross-sectional weights (_ub and _ui): would it be <br />f_indscub_xw or f_indscui_xw?<br />For wave 7 it would have the value of: g_indscui_xw (as there is no _ub version)<br />For wave 8 it would have the value of: h_indscui_xw (again, as there is no _ub version)<br />For wave 9 it would have the value of: i_indscui_xw (again, as there is no _ub version)</p>
<p>Then, turning to filling in the COVID-19 survey values of the new ‘weight’ variable it would be:<br />For wave 1 of the COVID-19 survey it would have the value of: ca_betaindin_xw <br />For wave 2 of the COVID-19 survey it would have the value of: cb_betaindin_xw <br />…<br />For wave 8 of the COVID-19 survey it would have the value of: ch_betaindin_xw<br />For wave 9 of the COVID-19 survey it would have the value of: ci_betaindin_xw</p>
<p>So, the Stata code would look something like this to give you an idea of what I mean:</p>
<p>gen weight = b_indscub_xw if mainstage_wave==2<br />replace weight = c_indscub_xw if mainstage_wave==3<br />replace weight = d_indscub_xw if mainstage_wave==4<br />…<br />replace weight = i_indscui_xw if mainstage_wave==9<br />replace weight = ca_betaindin_xw fi cv19survey_wave==1<br />…<br />replace weight = ch_betaindin_xw fi cv19survey_wave==8<br />replace weight = ci_betaindin_xw fi cv19survey_wave==9</p>
<p>So, it would be one new ‘weight’ variable, where each wave within each pidp had a weight value which corresponds to the cross-sectional weight for that wave.</p>
<p>- Is that the correct approach to take in to treat the data as repeated cross-section data and look at levels of mental health in each wave? <br />- Am I handling the COVID-19 weights correctly, and can I combine the Mainstage (waves 1-9) and Covid-19 surveys (waves 1-9) in this way?<br />- I’m not sure I fully understand the switch between _ub (waves 1-6 mainstage) and _ui (waves 7-9 mainstage). I can only use _ub up to wave 5 and only _ui from waves 7 to 9. Is it correct to take the approach I’ve outlined above, looking at _ui in some waves and _wub in others? <br />- Also, for wave 6, which self-completion cross-sectional weights should I use? The _ui or _ub?</p>
<p>I hope this makes sense and please do let me know if you require any further clarifications.</p>
<p>Best wishes,</p>
<p>James</p> Support #1937 (Resolved): "Inapplicable" responses for dweltyphttps://iserredex.essex.ac.uk/support/issues/19372023-07-10T14:08:37ZVikram Patil
<p>Hello, <br />My question is about the variable "dweltyp" in the hhsamp files. <br />- Why are so many households reported as "-8" or "inapplicable" from Wave 7 onwards in the Understanding Society (USoc) data? <br />- How is "dweltyp" different from the "hstype" variable that was included in the BHPS?<br />- Why are there no inapplicable responses prior to Wave 7? <br />Thanks!</p> Support #1869 (Resolved): 2021 calendar datahttps://iserredex.essex.ac.uk/support/issues/18692023-02-28T10:00:18ZMatthias Pierce
<p>Hi, do you have any further updates on when the 2021 calendar data will be available? Sorry to be persistnent, but we would like this data to complete our analysis!</p>
<p>Thanks</p> Support #1836 (Resolved): 2021 calendar datahttps://iserredex.essex.ac.uk/support/issues/18362023-01-09T11:39:56ZMatthias Pierce
<p>Hi, do you have an update on when the 2021 calendar data will be released? <br />Thanks</p> Support #1825 (Resolved): 2011 LSOAs for BHPS waveshttps://iserredex.essex.ac.uk/support/issues/18252022-12-06T18:01:17ZAlbert Wardalbert.ward@politics.ox.ac.uk
<p>Hi,</p>
<p>I was wondering whether you have any plans to make the 2011 LSOA file (<a class="external" href="https://beta.ukdataservice.ac.uk/datacatalogue/studies/study?id=7248&type=Data%20catalogue">https://beta.ukdataservice.ac.uk/datacatalogue/studies/study?id=7248&type=Data%20catalogue</a>) apply to the BHPS waves, in the same way that the 2001 LSOA file (<a class="external" href="https://beta.ukdataservice.ac.uk/datacatalogue/studies/study?id=6670&type=Data%20catalogue">https://beta.ukdataservice.ac.uk/datacatalogue/studies/study?id=6670&type=Data%20catalogue</a>) applies to the UKHLS waves of the survey.</p>
<p>I ask because we are only permitted to match one or the other to the Special Licence file, and therefore it is not possible to study the whole panel using the most up-to-date LSOAs.</p>
<p>Thanks,</p> Support #1814 (Resolved): "inapplicable" in sempderived variablehttps://iserredex.essex.ac.uk/support/issues/18142022-11-22T13:21:55ZClaire Wu
<p>Dear support team,</p>
<p>I have a question about the "semderived" variable in the Covid-19 study. I intend to create a binary variable that includes people who are in paid employment and people who are not employed. However, I am very confused as to what the "not applicable" category represents in this variable. Are these people not employed?</p>
<p>Thank you very much for your help!</p>
<p>Claire</p> Support #1723 (Resolved): 'Other qualifications' category in hiqual_dvhttps://iserredex.essex.ac.uk/support/issues/17232022-07-07T08:49:00ZAlbert Wardalbert.ward@politics.ox.ac.uk
<p>Hi,</p>
<p>I can't find any information on what exactly 'other qualifications' consists of in your hiqual_dv variable (<a class="external" href="https://www.understandingsociety.ac.uk/documentation/mainstage/dataset-documentation/variable/hiqual_dv">https://www.understandingsociety.ac.uk/documentation/mainstage/dataset-documentation/variable/hiqual_dv</a>)</p>
<p>Would you be able to provide a quick explanation or point me in the right direction?</p>
<p>Many thanks</p> Support #1641 (Resolved): 3 critical issues regarding COVID-19 UK Household Longitudinal Datasethttps://iserredex.essex.ac.uk/support/issues/16412022-01-28T04:24:00ZSeung Un Lee
<p>Dear UCL Institute of Education,</p>
<p>I am Seung Un Lee, currently studying Master's of Urban Economics at the National University of Singapore. I have a few critical issues regarding UK longitudinal studies - Covid19 dataset that your institution collects and provides and I was hoping if you could clarify it. Before that, I want to say thank you on behalf of our research team, with Associate Professor Kwan Ok Lee (profile) and Assistant Professor Michael Mai (profile), as we are using this dataset for our research in finding a correlation between Mental Health and Covid 19 National Lockdown.</p>
<p>The issues are as follows:</p>
<p>First, your team has collected respondents' various long-term health conditions. However, the responses seemed to be incompatible along the waves. I have attached a datasheet of the "Mentioned" response ratio by each wave. As you can see, Wave 2 and Wave 3 are way off the pattern. (The values were extracted from the raw dataset and cross-checked from the CLOSER discovery website.) It seems pretty clear that there is a problem with the values, so is there any reason behind this fluctuation? and how can we alter this value to make it consistent along the waves? Is there any way we can fix it to use it in our research?</p>
<p>Second, our team is using the dataset collected from Wave 1 (April 2020) to Wave 8 (March 2021), but we are hoping if you could clarify the exact date of when the survey started and ended. This is to match the period of National Lockdown that was imposed 3 times, including the most recent one in January, so that we can see how it affected the respondents' condition.</p>
<p>Third, we checked that there are two types of sampling weight, cX_betaindin_xw (cross-sectional) and cX_betaindin_lw (longitudinal). I tried to check the difference but the dictionary doesn't hold much information regarding these two variables. Can you please elaborate? We would be grateful if you could spare your time to give us a reply because the sampling weight issue is also another very important factor for our research and the dataset credibility.</p>
<p>Thank you again so much, and I will be looking forward to receiving any comments from you all! A bit late but, Happy New Year and hope you have a wonderful day :)<br />Sincerely, Seung Un.</p> Support #1600 (Resolved): 'Other' response in hsownd variablehttps://iserredex.essex.ac.uk/support/issues/16002021-10-26T11:22:42ZJames Morrison
<p>Hello,</p>
<p>I am working on a project looking at economic security across different housing tenures. Is there any more information on what kinds of living arrangements are covered in the 'other' response to the hsownd variable. I have consulted the questionnaire but the help section only includes information about shared ownership.</p>
<p>On an unrelated note, I have been having trouble accessing the support forum via the website and have had to do so via links in emails notifying me of responses to previous queries. I suspect this may be because the link directs users to 'iserswww.essex.ac.uk/support', whereas the forum now seems to be hosted at 'iserredex.essex.ac.uk/support'.</p>
<p>Thanks,<br />James</p> Support #1584 (Resolved): "inapplicable" end date of maternity leave & how to find out when leave...https://iserredex.essex.ac.uk/support/issues/15842021-09-17T17:25:08ZLaura Joneslaura.jones@nesta.og.uk
<p>Dear support team,<br />I am trying to create a variable measuring length of labour market interruption following childbirth in months.</p>
<p>One important component of this is knowing when maternity leaves end - variable w_matlvendm .<br />However, I find that a large chunk of my sample have w_matlvendm & w_matlvendy listed as inapplicable despite the fact that they have non-missing values on variables w_matlvstm and w_matlvsty (maternity leave start month and year).</p>
<p>On closer inspection I see that these are people who have a value of 3 on variable w_matlv (SPONTANEOUS Currently on maternity/paternity leave) - i.e. they are still on maternity leave at the time of the interview.<br />For some of these people w_jbstat maternity leave and so I am able to trace the end of their leave at the next wave using notempchk, empstendm etc.</p>
<p>However for a good proportion of them w_jbstat paid employment (ft/pt) and so, as far as I can tell there is no way of tracing the date of the end of their leave at the next interview.</p>
<p>Can I ask whether there is something I am missing in terms of finding out the end date of their leave/ when they returned to their jobs? Can you suggest any alternative methods?</p>
<p>I’m also curious as to why in some cases jbstat = paid employment when matlv currently on mat leave, whereas in others jbstat maternity leave.</p>
<p>many thanks in advance,<br />Laura</p> Support #1232 (Resolved): 2015 immigrant and ethnic minority boost and ward level datahttps://iserredex.essex.ac.uk/support/issues/12322019-08-23T12:08:25ZElena Pupazae.c.pupaza@lse.ac.uk
<p>To whom it may concern,</p>
<p>I don't seem to be able to find the 2015 Understanding Society immigrant and ethnic minority boost on the UK Data service as a separate dataset. Is it already merged and available with the main dataset or do I need special licence to access it?</p>
<p>Secondly, I am looking for finer geographical information for the immigrant and ethnic minority boost (ideally at the ward level) . Could you please advice me which of the special licence versions includes this information?</p>
<p>Thank you very much in advance!</p> Support #1197 (Resolved): 'inapplicable' category for jbft_dv variablehttps://iserredex.essex.ac.uk/support/issues/11972019-06-11T15:38:51Zfabiana macorfabianamacor@gmail.com
<p>Hi there</p>
<p>I have a query tangentially related to a previous issue (REF 1063) [[<a class="external" href="https://iserswww.essex.ac.uk/support/issues/1063">https://iserswww.essex.ac.uk/support/issues/1063</a>]], in relation to the jbft_dv.</p>
<p>jbft_dv has the following labels corresponding to 'missing' values (albeit different types of missing):</p>
<p>Label -> code<br />missing -> -9<br />inapplicable -> -8<br />proxy -> -7</p>
<p>The variable note states that "Inapplicable [-8] to proxy respondents due to missing information on overtime, and respondents who do not have a paid job", which I find confusing because there is already a code for proxy respondents [-7]. So, if proxy respondents don't have the information and this is coded as [-7], surely the [-8] code refers only to respondents who do not have a paid job?</p>
<p>Essentially I am trying to see which respondents shift from FT employment (wave 1) to non-employment (wave2), and I would like to double check that 'inapplicable' will capture all those that are not in employment (the variable of interest for wave 2). If this is not possible, how can the 'inapplicable' category be combined?</p>
<p>Many thanks in advance for your assistance</p>
<p>Fabiana</p> Support #992 (Resolved): Individual-level Nonresponse Adjustmenthttps://iserredex.essex.ac.uk/support/issues/9922018-07-01T12:41:43ZAnte Bab2242@cam.ac.uk
<p>Dear Sir or Madam,</p>
<p>I'm referring to the Understanding Society, Waves1-5, User Manual. According to the Manual, the derivation of the weighting adjustments for individual-level non-responses includes a logistic regressions as described as follows (p. 72):</p>
<p>"The predictors used in the models include all the predictors used for the household level nonresponse models and individual and household-level variables obtained from the household questionnaire, such as age and gender, marital and employment status, household size and presence of children in the household, as well as household expenditure on food and food outside, consideration of use of environmental energy, among others."</p>
<p>Is there a complete list that shows all the factors used in the regression? I'm asking because I would like to check whether the Big Five personality traits from Wave 3 are included (non-response might well depend on personality traits). I doubt that they are because they were only part of Wave 3 but I would like to double-check. Also, I'm interested whether education is accounted for.</p>
<p>Thank you very much for your help.</p>
<p>Best regards,<br />Ante</p> Support #945 (Resolved): total number of siblings of a given respondent (gender is irrelevant) https://iserredex.essex.ac.uk/support/issues/9452018-03-25T15:59:24ZNico Ochmannnico.ochmann@postgrad.manchester.ac.uk
<p>Dear Support Team,</p>
<p>I am looking for the total number of siblings of a given respondent (gender is irrelevant).</p>
<p>I could not find it, and I am not sure if there is information available to derive such a variable.</p>
<p>Your help is much appreciated.</p>
<p>Best wishes.</p>
<p>Nico</p> Support #877 (Resolved): weights for pooled cross-sections over waves (a)-(f) https://iserredex.essex.ac.uk/support/issues/8772017-11-10T14:55:16ZNico Ochmannnico.ochmann@postgrad.manchester.ac.uk
<p>Dear Peter, <br />I have a few more questions with regard to my original support <a class="issue tracker-3 status-5 priority-4 priority-default closed" title="Support: weights for pooled cross-sections over waves (a)-(f) (Closed)" href="https://iserredex.essex.ac.uk/support/issues/758">#758</a>. <br />1. When I generate my new weighting variable, newwgt, 600 observations in my estimation sample are assigned a zero meaning they are dropped. Why do you guys assign zero weights in w_indinus, w_indinub, etc. ? <br />2. The sample mean of my newwgt is 1.1 with a s.d. of 0.58. Does this look reasonable to you as in theory the mean should be 1. <br />3. Does USoc normalize weights so that N observations in weighted data equal N observations in unweighted data? It seems to me that you guys do that. <br />4. Given the normalization, how would I find the number of weighted and unweighted observations with Stata for my estimation sample?</p>