Question

For this forum, you are given five different study scenarios (A, B, C, D, and E). Your task is to examine each scenario and propose what source(s) of error exists and why. For each scenario, you should suggest the type(s) of error (e.g. random error, information bias, selection bias, and confounding) and why you think these errors are present in the given scenario.

Scenario A:

Stress is measured in a study based on a person's cortisol levels each day. However, cortisol levels vary dramatically throughout the day.

Scenario B:

A study was conducted to examine whether a woman's diet during pregnancy is associated with her child's IQ at age 7 years. When her child turns 7 years old, the researcher assesses his/her IQ and surveys the woman's diet during her pregnancy.

Scenario C:

A study finds a positive association between outdoor air pollution and asthma. Air pollution was measured by the weather center and asthma was measured by diagnosis.

Scenario D:

A study follows women over time to determine the relationship between depression and early death. 400 female psych patients were recruited from the local hospital and followed for 10 years. 50% of the study participants were lost over the 10-year period due to non-response.

Scenario E:

A study examined the relationship between weight in pounds and risk of heart attack. Heart attack was determined from medical records and weight was determined by participants' self-reports.

Answer 1

A. The error in this scenario is random error. Random errors in experimental measurements are caused by unknown and unpredictable changes in the expirement .As a result values may vary from one measurement to another. First case cortisol level may vary dramatically throughout the day

B Random errors may arise during the measurement of food-intake on a given day. random errors usually occur when experimenter's inability to take measurement in exactly same amount to get the exact number. Food intake may vary throughout the day so accurate measurement is not possible. During pregnancy, depending upon the trimester intake of food may vary

C. Confounding bias is the error in this scenario. Confounding variable is an extraneous variable which affects both independent and dependent variable of the study and thus the study results . Because both asthamatic attack and air pollution have strong temporal cycles defined by weather , day of week, seasonal and long-term trends

D.Selection bias can result, when selection of study subjects or their likelihood of being retained in the study is affected. This study is conducted for a period of 10 years and the researcher had lost 50% of study samples , hence the sample obtained is not representative of the population intended to be analyzed.

E. Information bias occurs when there is distortion in the association of key variables caused by a lack of accurate measurements of key variables. Weight is one of the key variable in the study to find out relationship betwen heartattack and weight, which is not measured using reliable weighing scales. So the error is information bias