No menu items!

Undercoverage

-

When drawing a sample from a population, if a certain section of the population is excluded from the sample then it is called undercoverage. Undercoverage bias introduces error into our study because the sample chosen is not truly representative of the population.

Some examples of undercoverage bias/error:

  1. Suppose that you wish to conduct a study to find out the average income of a population. You may randomly collect information from people who are having dinner at different restaurants and consider that as your sample. But the results of this study will not be accurate. This is because your sample excludes all those people who do not like to eat out, it excludes people who may not be able to afford having dinner at restaurants, etc. Thus your sampling method has a systematic bias and is introducing an undercoverage error in your study.
  2. Suppose that you try to find out the political preferences of voters by means of an online poll. The data obtained from such a method will not be representative of the society as a whole. Some people may not have internet access or may simply be tech savvy. Also since the website depends on voluntary responses it is more likely to attract political enthusiasts who are passionate about their positions. Thus this method systematically excludes those who may be generally less enthusiastic about political matters.
  3. Suppose that you wish to study the efficacy of a particular teaching method and its impact on students’ performance. When selecting the students who will be participating in the study it is important that the researcher chooses students of different academic ability. If the researcher only chooses students with good academic ability, the study will be biased in positively in favour of the teaching method. The results of such as student cannot be objective.
It is important to remove undercoverage bias from our sample.

Reasons for undercoverage bias:

  1. Bias of the Researcher – If the researcher has already made his mind about the conclusions of the study then he may try to manipulate the sample to exclude those units which may lead to conclusions opposite to the one he favours. This will introduce undercoverage bias into the study.
  2. Incorrect sampling methodology – This may happen if people asked to fill a questionnaire but the questions are available in only one language. This creates a bias against people who do not read that language and who may therefore not fill the form.

How to eliminate undercoverage bias:

The sample must be chosen in a random manner so that it is unaffected by the bias of the researcher. The sampling methodology must be robust to make sure that all groups in a population have equal chances of being included as a unit in the sample.

Hey 👋

I'm currently pursuing a Ph.D. in Maths. Prior to this, I completed my master's in Maths & bachelors in Statistics.

I created this website for explaining maths and statistics concepts in the simplest possible manner.

If you've found value from reading my content, feel free to support me in even the smallest way you can.



Share this article

Recent posts

Popular categories

Recent comments