Stats Question (1 Viewer)

[davidgoes4wce]

The administration of 'Knowitall University' want to know what proportion of their students are satisfied of their students with the library facilities. The library staff are keen to help. During the 1st week of the exam period they choose 200 students who are in the library and ask them to fill in a questionnaire on satisfaction with the library. It we wanted to draw conclusions about all students who used the library in the first week of the exam period, the sample selection was:

*Biased
*Unbiased
*Can't tell

 

[InteGrand]

The administration of 'Knowitall University' want to know what proportion of their students are satisfied of their students with the library facilities. The library staff are keen to help. During the 1st week of the exam period they choose 200 students who are in the library and ask them to fill in a questionnaire on satisfaction with the library. It we wanted to draw conclusions about all students who used the library in the first week of the exam period, the sample selection was:

*Biased
*Unbiased
*Can't tell

I don't think we can tell, because we aren't told if the surveyed students were selected randomly or not. We aren't given details of the survey sampling method used.

 

[davidgoes4wce]

Answer: The students were selected by the library staff. The staff may have subconsciously avoided selecting students who look dissatisified.

 

[InteGrand]

Answer: The students were selected by the library staff. The staff may have subconsciously avoided selecting students who look dissatisified.

Yeah that's what I thought. So we can't conclude whether or not it is biased, because we aren't told how the selection was actually conducted (it may have been random using a well-known probability sampling method (http://stattrek.com/survey-research/sampling-methods.aspx?Tutorial=AP )), but it may also have not been).

 

[davidgoes4wce]

the answer was "Probably Biased"

but some of the wording from this online quiz are very ambiguous.

 

[InteGrand]

the answer was "Probably Biased"

but some of the wording from this online quiz are very ambiguous.

That wasn't even an option from the three options you gave.

 

[davidgoes4wce]

That wasn't even an option from the three options you gave.

My bad the choices were:

Probably Biased, Probably Unbiased, We Can't tell

 

[davidgoes4wce]

Reading the question again, I say that the population was biased. I'm reading it as the 'library staff were keen to help' students. If it was an unbiased sample the librarians would be doing the research in a neutral manner. I mean I often think of lecturers when in my undergraduate degree giving out 'Evaluation forms' about their teaching methods during a semester. You want to give out the forms based on how you have taught the unit the whole semester not just winning over students in the last week of a semester. It distorts the study, it defeats the purpose of proving a 'neutral' perspective on the sample.

 

[GoldyOrNugget]

This is such a bad question... I'd probably say probably unbiased because I'm guessing the question-setters aren't expecting you to read too deeply into the motivations of the librarians. I think they want you to decide if the experimental setup is fundamentally unbiased, which it's not. I think the random selection of students is meant to be implicit in the question.