https://iserredex.essex.ac.uk/support/https://iserredex.essex.ac.uk/support/support/favicon.ico?15995719382019-04-17T08:40:54ZUnderstanding Society User SupportUnderstanding Society User Support - Support #1185: Linking all waves of BHPS and UKHLS: Inconsistencies?https://iserredex.essex.ac.uk/support/issues/1185?journal_id=43852019-04-17T08:40:54ZAlita Nandi
<ul><li><strong>Status</strong> changed from <i>New</i> to <i>Feedback</i></li><li><strong>Assignee</strong> changed from <i>Alita Nandi</i> to <i>Nicole Schwitter</i></li><li><strong>% Done</strong> changed from <i>0</i> to <i>80</i></li><li><strong>Private</strong> changed from <i>Yes</i> to <i>No</i></li></ul><p>Hi Nicole,</p>
<p>Without looking at your code I cannot comment but I can say that when I appended 25 waves I too find similar inconsistencies. The reason you are getting these inconsistencies is because of proxy interviews. For these cases sex is coded as -7. We will look into why sex=-7 for proxy interviews as the missing value code of -7 is reserved for cases where the information is missing for proxy interviews - and that is not the case for sex.</p>
<p>Detailed answer:</p>
<p>After creating the long format file of 25 waves, I produced the mean of sex, sex_mean1 and counted mismatches of this mean with individual wave specific value of sex. I too found 15458 inconsistencies</p>
<p>bys pidp: egen sex_mean1=mean(sex)<br />cou if sex~=sex_mean1</p>
<p>Then I recoded proxies to missing and repeated the exercise and found 2545 mismatches.<br />recode sex -7=.<br />bys pidp: egen sex_mean2=mean(sex)<br />cou if sex~=sex_mean2</p>
<p>And when I restricted this to only those cases where sex is not missing the number of mismatches goes down to 492.</p>
<p>cou if sex~=sex_mean2 & sex<.</p>
<p>Best wishes,<br />Understanding Society User Support Team</p> Understanding Society User Support - Support #1185: Linking all waves of BHPS and UKHLS: Inconsistencies?https://iserredex.essex.ac.uk/support/issues/1185?journal_id=43872019-04-17T09:02:39ZAlita Nandi
<ul></ul><p>Also, sex_dv is only available for the 8 UKHLS waves and it does not have the same problem, that is, there is a valid value even for proxy respondents.</p> Understanding Society User Support - Support #1185: Linking all waves of BHPS and UKHLS: Inconsistencies?https://iserredex.essex.ac.uk/support/issues/1185?journal_id=43882019-04-17T09:19:24ZNicole Schwitter
<ul></ul><p>Thank you Alita!<br />This is very helpful as I was mostly wondering whether I did something wrong in the matching process or whether it is something inherent in the data, but this explains the problem.</p> Understanding Society User Support - Support #1185: Linking all waves of BHPS and UKHLS: Inconsistencies?https://iserredex.essex.ac.uk/support/issues/1185?journal_id=59232021-03-02T15:56:45ZUnderstanding Society User Support Teamusersupport@understandingsociety.ac.uk
<ul><li><strong>Status</strong> changed from <i>Feedback</i> to <i>Resolved</i></li><li><strong>Assignee</strong> deleted (<del><i>Nicole Schwitter</i></del>)</li><li><strong>% Done</strong> changed from <i>80</i> to <i>100</i></li></ul>