Research Methods and Statistics

The methodology of the most recent SONA study that I have been a participant for.

The most recent SONA study that I have taken part in was researching how much alcohol students consume and the reasons behind the drinking.

To research how much alcohol I had consumed they asked to to report how much alcohol I had consumed in the last four weeks (self-report method). For this I had to write down how much alcohol was consumed on which days and if it was a special occassion or not. This is not a very reliable way of collecting this information as I did not know I would have to do this so it was difficult to remember when I drank and exactly how much, and I expect this would have been a more unreliable measure for someone who drinks more frequently than I do.

To research the reasons for drinking and my mood when I was drinking I was asked to complete (many) closed question questionnaires. These were often repetitive and, in my opinion missed some reasons that people have for drinking. It concentrated more on the social pressures that may cause people to feel pressured into drinking. This series of questionnaires ended with a likert scale measuring how much I thought alcohol helped or hindered certain areas of life such as university work and future job prospects.

Overall this study is one of the simpler SONA experiments I have taken part in and found it a much easier one to concentrate on despite the repetitiveness (which leads to boredom), however I feel if they asked participants to keep track of drinking habits for a certain amount of time then their data may be more reliable than the participants trying to remember what they have had.


Single Post Navigation

15 thoughts on “The methodology of the most recent SONA study that I have been a participant for.

  1. I like your blog and personally i myself have participated in the same, or a very similar study, and i agree with your comments about the reliability, however i don’t believe telling people to count how much they drank in a week, maybe it would work for a normal sample, but i think for students it i not reliable, as i personally know some people that cannot tell what they have had to drink after a certain point, also telling people in advance might lead them to drink less during that particular time, to be perceived in a better light. However to conclude, i agree with you that this is a very simple study, maybe even too simple, and it does not really provide valid measures, which is one of the biggest issues with using self-report measures.

  2. Pingback: Homework for my TA week 8 « Nirapsy

  3. Questionnaire’s are one of my preferred SONA studies to take part in, simply because they are not as boring as the majority. Order effects such as boredom and fatigue are big problem in studies that have reaction time tests or a what appears to be a never ending list of question about something utterly mind numbing such as identifying the correct shapes rapidly. I think that studies lasting over an hour should be worth more credits than they are, especially when its spending an hour staring at a screen flashing objects at you. The methodology should try and be made a little more entertaining, this could be done by trying to make the study more life like. Virtual reality experiments offer more ‘excitement’ to a setting but even then the reality is still not offering total ecological validity, plus problems that occur from watching the tv/computer screen for too long without a break maybe a problem.


    • I do not agree with you that studies lasting over an hour should be worth more credits but I do think that real breaks should be set. This is because repetitive tasks and tasks involving staring at computer screens and pushimg buttons can, in my experience, be harmful. Unfortunately the majority of the SONA experiments are computer based and looking for too long causes me to have headaches and it hurts my eyes, and even when the study says to take a break often I just carry on as the experimenter is in the room so I feel that I should carry on as I do not want to be percieved as lazy or something. for this reason I feel that real breaks should HAVE to be given to participants and this should be monitored. As well as protecting participants from harm, this would also stop the participants from feeling as fatigued as they would get a break to re-energise.

  4. Whilst I agree that it is possible your responses to the self report sections of the study may not have been as accurate as they could be I would argue that they are still better than they would be if you had been told in advance as self monitoring has been shown to affect target behaviours (Broden et al. 1970). In this experiment if participants were told to pay extra attention to how much they drink during the week those who drink a lot may notice this and as a result drink less, biasing the results. Alterntively they may just deliberately misreport the amount they drink due to social desirability bias, a source of error which King (2000) describes as all too often overlooked or ignored. And of course inevitably some participants will still make mistakes so unfortunately that source of error would still not be enitrely removed from the equation. All in all I think the researcher chose the best method of gathering that data given the subject matter although had they been measuring something that people have a neutral opinion on (e.g. number of emails sent) perhaps your method would have provided more accurate and reliable results.

  5. Pingback: Comments for my TA – 14.03.12 « groundblogday20

  6. Pingback: Homework for TA « Between the lines…

  7. Pingback: Comments 14/03/2012 « statsblog2011

  8. Hi, I like how you explained the experiment well in a short and simple way. 😀
    I have participated in a same/similar study about drinking, and I totally agree with the repetitiveness, which I remember some of the questions have been asked for 3 times,and everytime I saw the same questions appear again, I would get rather confused(probably because I was guessing why and getting slightly annoyed)
    the previous comments mentioned that self report may not be as accurate however I think self report is quite accurate, at least when compare to interviews self report should get more accurate results. I would like to act good in front of someone but not to when completing an anonymous questionnaire. “Accuracy of Self-Reported Drinking: Observational Verification of ‘Last Occasion’ Drink Estimates of Young Adults” (Northcote & Livingston,2011) I found this study is a similar search but was done in Australia, it is suggested that these recall methods in self-report surveys are potentially reasonably accurate measures of actual drinking levels for light to moderate drinkers. And I agree with your point that the research should keep on track with Participants’ drinking habits rather than asking the previous week/month drinking consumption only because people may drink more this month, less next month, it could varied over time. For example, I could drink a lot on one night and don’t drink again for 1/2 months time. Also, getting the participants to do the questionnaire online maybe easier too, since people maybe more willing to do it and it makes follow up questionnaire more possible.

  9. I agree with you. I had to do one very similar about my alcohol intake, and also about whether I currently needed a drink. In fact, there were several questionnaires to fill in, and three of them I had to complete twice (once and the beginning and once at the end), and I couldn’t even see the logic since one of the repeated questionnaires was demographics. I was still 20 and female, even though I did feel like I’d lost part of my life!

    But the actual questionnaires themselves – I had no idea how much alcohol I’d consumed, and working out the units was difficult because of this. I also question how honest people would be on some of them – do you really want to tell a complete stranger that you’re reliant on alcohol and craving a drink? Also, because it only asks within the last month, it’s not entirely accurate. For example, it had just been the end of exams so I’d gone out a few times and drank too much, whereas on a normal month that would have been a lot less. But although repetitive like you said, I find I answer a lot better since I don’t lose concentration or get sleepy/bored like in computer based tasks.

  10. Pingback: Homework for my TA « statsbloggy

  11. Pingback: Comments for TA due 14/3/2012 :) | The Blog :) by poey

  12. You have a well written blog and I enjoyed reading it.
    Reports regarding alcohol especially can lead to being very unreliable, especially within some target populations. When asked about how much I consume in a typical week, depending on my interpretation can vary dramatically. Whilst at university, pretty high, but whilst I am at home I drink a lot less. Therefore if I took my average over the year or a couple of months this could produce very different and unrealistic results. This is more a problem with the research questions given and poorly designed studies as it is not specified or given a time period. The other problem with self reports and questionnaires is the possibility of people lying, Whether this be over or under exaggeration. Closed questionnaires are a lot easier to analyse so preferable from the researchers point of view however these come with there own limitations. However, in some studies although limiting closed questions will give the best results as through open questions you may not actually get the question you wanted answered.

  13. Pingback: Comments for 18/4/12 « psud43

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: