Health related surveys: sampling weights
n a previous article in this statistical primer series, we introduced commonly used sampling methods in health-related surveys. They include simple random sampling, stratified random sampling, and cluster sampling. If you have not read that primer already, or feel you need a refresher on the topic,...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
New Zealand Medical Student Journal Society
2024-12-01
|
| Series: | New Zealand Medical Student Journal |
| Online Access: | https://doi.org/10.57129/001c.127913 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832586228043612160 |
|---|---|
| author | Ari Samaranayaka Robin Turner Claire Cameron |
| author_facet | Ari Samaranayaka Robin Turner Claire Cameron |
| author_sort | Ari Samaranayaka |
| collection | DOAJ |
| description | n a previous article in this statistical primer series, we introduced commonly used sampling methods in health-related surveys. They include simple random sampling, stratified random sampling, and cluster sampling. If you have not read that primer already, or feel you need a refresher on the topic, we suggest reading it before reading this article. A single complex survey can include a combination of the above sampling methods. All of these sampling methods aim to
achieve a sample that is representative of the underlying study population. Simple random sampling achieves this representativeness by ensuring (as much as possible) that everyone in the population has the same chance of being selected. The assumption is made that observations from di-erent participants are independent of each other, and each participant represents the same number of individuals in the
population, therefore those observations contribute equally to the findings. These assumptions mean that basic statistical methods can be used to analyse the data from a simple random sample. As you can probably imagine, there are many situations where this sampling method is not possible or appropriate. Examples of circumstances where simple random samples are not possible include, but are not limited to: instances where we are not able to list the people in the
population (non-availability of a sampling frame that includes everyone), which is needed to choose people at random; when participation of individuals who have been sampled are di-erent between its subgroups; and when the number of people selected from minority groups is too small. |
| format | Article |
| id | doaj-art-d2abec1c23b34d82bab5fcb4860bebbc |
| institution | Kabale University |
| issn | 1176-5178 1179-3597 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | New Zealand Medical Student Journal Society |
| record_format | Article |
| series | New Zealand Medical Student Journal |
| spelling | doaj-art-d2abec1c23b34d82bab5fcb4860bebbc2025-01-26T08:05:51ZengNew Zealand Medical Student Journal SocietyNew Zealand Medical Student Journal1176-51781179-35972024-12-01038Health related surveys: sampling weightsAri SamaranayakaRobin TurnerClaire Cameronn a previous article in this statistical primer series, we introduced commonly used sampling methods in health-related surveys. They include simple random sampling, stratified random sampling, and cluster sampling. If you have not read that primer already, or feel you need a refresher on the topic, we suggest reading it before reading this article. A single complex survey can include a combination of the above sampling methods. All of these sampling methods aim to achieve a sample that is representative of the underlying study population. Simple random sampling achieves this representativeness by ensuring (as much as possible) that everyone in the population has the same chance of being selected. The assumption is made that observations from di-erent participants are independent of each other, and each participant represents the same number of individuals in the population, therefore those observations contribute equally to the findings. These assumptions mean that basic statistical methods can be used to analyse the data from a simple random sample. As you can probably imagine, there are many situations where this sampling method is not possible or appropriate. Examples of circumstances where simple random samples are not possible include, but are not limited to: instances where we are not able to list the people in the population (non-availability of a sampling frame that includes everyone), which is needed to choose people at random; when participation of individuals who have been sampled are di-erent between its subgroups; and when the number of people selected from minority groups is too small.https://doi.org/10.57129/001c.127913 |
| spellingShingle | Ari Samaranayaka Robin Turner Claire Cameron Health related surveys: sampling weights New Zealand Medical Student Journal |
| title | Health related surveys: sampling weights |
| title_full | Health related surveys: sampling weights |
| title_fullStr | Health related surveys: sampling weights |
| title_full_unstemmed | Health related surveys: sampling weights |
| title_short | Health related surveys: sampling weights |
| title_sort | health related surveys sampling weights |
| url | https://doi.org/10.57129/001c.127913 |
| work_keys_str_mv | AT arisamaranayaka healthrelatedsurveyssamplingweights AT robinturner healthrelatedsurveyssamplingweights AT clairecameron healthrelatedsurveyssamplingweights |