QUESTION 1: (10 Marks)
Read the journal article “Survey subjects and the quality of health surveys”. This was given to you as a handout in the first lecture. It is also available on CANVAS. Answer the following questions:
a. What is the difference between a probability sample and a non-probability sample? Give examples. What is the advantage of using a probability sample? (1 + 3 = 4 marks)
A probability sampling procedure involves the selection of a probability which has a known, zero probability of selection of subjects from those eligible. For example, a population of 100 people, each person would have odds of 1 out of 100 of being chosen. In contrast, Non-Probability procedures give growth to the selection of non-probability samples which have an unknown, non-zero probability from a selection of subject from those eligible. For example, a person from the sample might have a better chance of being chosen if they live close to the researcher or have access to a computer.
Using a probability sampling enables the researcher the chance to create a sample that truthfully represents the population. This way, many selection biases are avoided, and the statistical theory can be used to derive properties of estimators such as prevalence’s, odds ratios and relative risks.
b. Using your own example, distinguish between a reference population and a source population (2 marks).
A reference population is a representative sample of individuals used to establish norms for reference ranges. Examples include, lawyers, people who smoke or people who meditate. However, because logistically it is impossible to include all members of a reference population, researcher will resort to using a source population. A source population is more restrictive when compared to a reference population. For example, a source population would include all people who live in Melbourne that meditate once a day or people who live in Melbourne and smoke a packet a cigarettes in a week.
c. What is a sample frame? (1 mark).
A sample frame, which is also known as a listing, is constructed from a source population. It is known as a list of all the units within the population of interest where all research findings is applied to the population defined by the sampling frame. The sample frame is defined with explicit inclusions and exclusions. For instance, the selection of RMIT University students may be confined to those who live within a 10km radius from the campus in 2018.
d. Name and explain the sources of error that affect the quality of health surveys indicating which errors are quantifiable and which are not quantifiable (3 marks).
There are two sources of errors that can affect the quality of health surveys, Selection error and Response error. A Selection error occurs when the source population is not representative of the reference population or the sampling frame is not representative of the source population. A response error generally occurs when sample study is not representative of the sampling frame. From a random selection sample, participants may choose not to participate, therefore a final sample may not be representative of the sampling frame. This type of error is not quantifiable.
Selection and Response errors are both not quantifiable; therefore they violate the assumption that each member of the source or reference population has the same, known probability of selection into the sample. When assumption is violated, the probability of selection is either zero or unknown. Non-quantifiable errors are typically ignored, which biases any conclusions that can be made from a study.