fredag 29 november 2013

Theme 4: Quantitative research

I chose the paper Social Networking Sites: Their Users and Social Implications — A Longitudinal Study by Petter, B. Brandtzæg (2012) 
1Which quantitative method or methods are used in the paper? Which are the benefits and limitations of using these methods?
The paper is using a longitudinal study that consisted of three survey waves. They collected data from a representative sample of 2,000 online users, aged 15–75 years, in Norway from 2008 to 2010.  The response rate was 71% in the first survey, 69% in the second and 35% in the last one. According to the writers the dropout rate is expected in longitudinal surveys.
It is always hard to know whom you are actually reaching with your survey. In the table in the paper it seems that they had a pretty even distribution over the different age groups and genders though. And another benefit is that Norway has an ideal population for studying the social impact of SNSs, given its exceptionally high Internet and SNSs connectivity (Internet World Statistics, 2010).

2What did you learn about quantitative methods from reading the paper?
That the dropout rate is so high, but still expected.

3Which are the main methodological problems of the study? How could the use of the quantitative method or methods have been improved?
The main problem I see is how they measure loneliness.
Five items from the Revised UCLA Loneliness Scale were included:
1) I feel in tune with the people around me,
2) No one really knows me well,
3) There are people I can turn to,
4) I feel left out,
5) I have difficulties in making friends.
All items were rated on a 4-point scale, ranging from (1) ‘‘Never’’ to (4) ‘‘Always’’.
Even though this seems to be a method that has been used before I don’t believe you can measure loneliness so easy. Loneliness is different to all people, and outing a number on your loneliness seems too simple.
“Physical Activity, Stress, and Self-Reported Upper Respiratory Tract Infection”.
They selected the participant’s randomly from the Swedish Population Registry. 5000 men and women aged 20-70 years were invited to participate in the study. A web questionnaire via email was sent out and then 5 follow up questionnaires. Out of the 5000 invited, 1111 people followed through the whole process, finishing all 5 questionnaires.
In quantitative research it is good that you reach a big number of people in a small amount of time, but it can be very hard to know whom you reach. We heard at the lecture the example of a couple of students researching iPad use. They sent out the questionnaire in an online forum, where all the members are men. The study then showed that 90% of iPad users are men, which of course isn’t true. But it was true in that study since all the participants in the study were men.  

Qualitative and quantitative research can often complement each other. Usually you cannot generalize conclusions with qualitative research, but you can originate hypotheses whose general validity can be verified for a large number of cases with quantitative method.
Examples of qualitative methods are surveys with open answers and interviews. A benefit of qualitative research is that you can get a deeper understanding for an issue, but it is usually very time consuming.




1 kommentar:

  1. That is a very good point about the target group when doing a study. I guess that is a very important part of the challenge for the researcher to make sure that the people answering your question is representative of everybody. I mean you cannot post your questions on Facebook, and expect to receive ideas from people who do not like Facebook, as they will likely not be there at all. Your target group must be equally important as eliminating disturbing variables from your test case.

    SvaraRadera