<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" version="2.0">
  <channel>
    <title>topic Re: PROC SURVEYLOGISTIC with clustered/correlated data in Statistical Procedures</title>
    <link>https://communities.sas.com/t5/Statistical-Procedures/PROC-SURVEYLOGISTIC-with-clustered-correlated-data/m-p/956087#M47835</link>
    <description>&lt;P&gt;The data comes from a large international organization and they did not publish their methodology, but shared the data publicly with weights that adjust for over- /under-sampling by age, education, income, etc. . The survey was implemented in several countries and the weights try to ensure that the data from each country are representative of that country. This is all I know. I am aware that it is likely not possible to appropriately estimate the variance and std errors using the available info. I do not yet have a model as I am in the process of determining which method and sas procedure to use. The data is clustered at the country level (i.e, respondents from the same country are more likely to be similar), I assume, hence the need to take clustering into consideration when fitting the model.&lt;/P&gt;</description>
    <pubDate>Tue, 14 Jan 2025 16:52:13 GMT</pubDate>
    <dc:creator>K_S</dc:creator>
    <dc:date>2025-01-14T16:52:13Z</dc:date>
    <item>
      <title>PROC SURVEYLOGISTIC with clustered/correlated data</title>
      <link>https://communities.sas.com/t5/Statistical-Procedures/PROC-SURVEYLOGISTIC-with-clustered-correlated-data/m-p/955908#M47824</link>
      <description>&lt;P&gt;I am analyzing data from a survey that was implemented in multiple countries. I am working with the assumption that the responses from subjects from the same country are more likely to be similar (within country correlation). How do I account for this in proc surveylogistic? Is it possible to account for this in proc surveylogistic?&lt;/P&gt;</description>
      <pubDate>Mon, 13 Jan 2025 13:49:08 GMT</pubDate>
      <guid>https://communities.sas.com/t5/Statistical-Procedures/PROC-SURVEYLOGISTIC-with-clustered-correlated-data/m-p/955908#M47824</guid>
      <dc:creator>K_S</dc:creator>
      <dc:date>2025-01-13T13:49:08Z</dc:date>
    </item>
    <item>
      <title>Re: PROC SURVEYLOGISTIC with clustered/correlated data</title>
      <link>https://communities.sas.com/t5/Statistical-Procedures/PROC-SURVEYLOGISTIC-with-clustered-correlated-data/m-p/956051#M47831</link>
      <description>&lt;P&gt;What is the sampling design you are working with and what is your current SURVEYLOGISTIC code?&lt;/P&gt;</description>
      <pubDate>Tue, 14 Jan 2025 13:36:12 GMT</pubDate>
      <guid>https://communities.sas.com/t5/Statistical-Procedures/PROC-SURVEYLOGISTIC-with-clustered-correlated-data/m-p/956051#M47831</guid>
      <dc:creator>SAS_Rob</dc:creator>
      <dc:date>2025-01-14T13:36:12Z</dc:date>
    </item>
    <item>
      <title>Re: PROC SURVEYLOGISTIC with clustered/correlated data</title>
      <link>https://communities.sas.com/t5/Statistical-Procedures/PROC-SURVEYLOGISTIC-with-clustered-correlated-data/m-p/956087#M47835</link>
      <description>&lt;P&gt;The data comes from a large international organization and they did not publish their methodology, but shared the data publicly with weights that adjust for over- /under-sampling by age, education, income, etc. . The survey was implemented in several countries and the weights try to ensure that the data from each country are representative of that country. This is all I know. I am aware that it is likely not possible to appropriately estimate the variance and std errors using the available info. I do not yet have a model as I am in the process of determining which method and sas procedure to use. The data is clustered at the country level (i.e, respondents from the same country are more likely to be similar), I assume, hence the need to take clustering into consideration when fitting the model.&lt;/P&gt;</description>
      <pubDate>Tue, 14 Jan 2025 16:52:13 GMT</pubDate>
      <guid>https://communities.sas.com/t5/Statistical-Procedures/PROC-SURVEYLOGISTIC-with-clustered-correlated-data/m-p/956087#M47835</guid>
      <dc:creator>K_S</dc:creator>
      <dc:date>2025-01-14T16:52:13Z</dc:date>
    </item>
    <item>
      <title>Re: PROC SURVEYLOGISTIC with clustered/correlated data</title>
      <link>https://communities.sas.com/t5/Statistical-Procedures/PROC-SURVEYLOGISTIC-with-clustered-correlated-data/m-p/956089#M47836</link>
      <description>Given the information you shared, it *might* be appropriate to use the CLUSTER statement.   Not knowing the survey design makes it difficult to know for sure.</description>
      <pubDate>Tue, 14 Jan 2025 17:24:05 GMT</pubDate>
      <guid>https://communities.sas.com/t5/Statistical-Procedures/PROC-SURVEYLOGISTIC-with-clustered-correlated-data/m-p/956089#M47836</guid>
      <dc:creator>SAS_Rob</dc:creator>
      <dc:date>2025-01-14T17:24:05Z</dc:date>
    </item>
  </channel>
</rss>

