Understanding Society User Support: Issueshttps://iserredex.essex.ac.uk/support/https://iserredex.essex.ac.uk/support/support/favicon.ico?15995719382016-03-17T09:31:40ZUnderstanding Society User Support
Redmine Understanding Society User Support - Support #531 (Closed): Context of blood samplinghttps://iserredex.essex.ac.uk/support/issues/5312016-03-17T09:31:40ZGaelle Albertus
<p>Hi,</p>
<p>I am working on the Nurse Health Assessment database (nurse and blood) combined to the full main interview. I only have included participants from wave 2.<br />I would like to analyze biomarkers and I need to adjust for the time of day the blood was taken.<br />I searched for such variable and I think the most appropriate variable is b_strtnurhh. The point is I obtained some people who have started the interview at 1 AM. Among 9,884 individuals, there are 73 in this case.<br />Here, there is STATA results:</p>
<p><code>table(b_strtnurhh, useNA="always")</code></p>
<pre><code><code>1 2 3 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 NA <br />73 1 1 2 1 3 47 566 1360 1076 583 610 837 725 750 792 1134 1037 258 27 1 0</code></code></pre>
<p>Should I identify these cases as missing values or is it a mistake and I can correct them by 1 PM ?<br />Thank you for helping me.</p>
<p>Gaelle</p>