Question

A statistics student who is curious about the relationship between the amount of time students spend on social networking sites and their performance at school decides to conduct a survey. Three research strategies for collecting data are described below. In each, name and discuss the sampling method proposed and any bias you might expect.

He randomly samples 300 students from the study's population, gives them the survey, asks them to fill it out and bring it back the next day. (1 mark).

He gives out the survey only to his friends, and makes sure each one of them fills out the survey. (1 mark).

He posts a link to an online survey on his Facebook wall and asks his friends to fill out the survey. (1 mark).

Victoria University has 1200 students undertaking the unit Introduction to Marketing this semester across three campuses with 900 students enrolled at the Melbourne City Campus, 200 students at the Sydney Campus and 100 students at the Sunway Campus. All lecture notes, teaching and supporting materials are uploaded on the VUCollaborate. The lectures are conducted by three different local lecturers at these campuses. The researcher wants to conduct a survey of 300 students about how satisfied the students are with the unit, and he believes that the campus a student is enrolled at might affect the student’s overall satisfaction with the unit.

Suggest the most appropriate sampling strategy for carrying out this study and any bias you might expect from it. .

Answer 1

1.1) He randomly samples 300 students from the study's population, gives them the survey, asks them to fill it out and bring it back the next day

The sampling method proposed is simple random sample

Bias you might expect : Non-response bias might be expected because he has given a very short duration in order to fill the survey.

1.2)He gives out the survey only to his friends, and makes sure each one of them fills out the survey

The sampling method proposed is convenience sample

Bias you might expect : This type of sampling may suffer from selection bias as the small group of friends that he chooses to fill the survey might not be representative of the entire population. Along with that, Non-response bias may creep in as some of his friends might not answer the survey.

1.3) He posts a link to an online survey on his Facebook wall and asks his friends to fill out the survey.

The sampling method proposed is convenience sample

Bias you might expect :This type of sampling may suffer from selection bias as the small group of friends that he chooses to fill the survey might not be representative of the entire population. Along with that, Non-response bias may creep in as some of his friends might not answer the survey.

Answer 2) The researcher can use proportional stratified random sample. Out of the total 1200 students that are enrolled in Introduction to Marketing, 900 (75%) of students are enrolled at the Melbourne City Campus, 200 (16.6%) students at the Sydney Campus and 100 (8.3%) students at the Sunway Campus.

Since the researcher has to take a sample of 300 students, he can divide the number of students that he takes from each campus based on the basis of total population that each campus has. For example, out of 300 samples needed for the study, 225 (75% of 300) can be taken from Melbourne City Campus, 50 (around 16.6 %) taken from Sydney Campus and the remaining 25 taken from Sunway Campus. This type of sampling may require more efforts than the simple random sampling.