Do you need statistics to understand your data?

Hmm, I personally think this week’s topic is quite a tricky one! Of course, statistics are very important and they do help us to understand data, but do we NEED them?

In terms of Psychology, the main purpose of the statistics is to show us whether or not our experiment has been successful… so whether or not our theory is worth mentioning. With the stats, we produce graphs so we can show off what we have found out from our experiment, we can use line graphs, bar charts, box plots, histograms or scatter graphs depending on what kind of data we have been dealing with (Ooo would you look at that, I was listening in week 2 small groups ;-)). With statistical graphs and tables everything is laid out nice and neatly infront of us, we can see correlation in the data, we can easily find averages, we can make sensible comparisons etc. The statistics basically allow us to make sense of theories.

…But, isn’t this only really relevant in terms of quantitative data?

 Statistics don’t really play a huge part with qualitative research. Qualitative researchers are more interested in why things are the way they are, rather than how they are. The main focus is people’s behaviours, attitudes and lifestyle. Interviews, emails, notes, case studies, photos, videos (and all that jazz ) are used for qualitative research… Rather than “yes or no” answer questions on questionnaires, which quantitative researchers like to use. So because qualitative data isn’t just simple numbers that can be added up (like “yes or no”, “male or female” answers on a questionnaire) it doesn’t always tend to end up in statistics. So obviously, statistics isn’t needed in understanding qualitative data.

So, I guess the answer to this question is.. No, We don’t need statistics to understand data. It sure does help us in terms of quantitative data, but not with qualitative. As I don’t consider myself to be the best statistician, I personally prefer the way qualitative researchers work. To me it feels like I’m actually getting to the point of the problem being investigated, rather than just using a load of numbers and sums to come out with one final figure which is supposed to be the answer to my problem!!

Advertisements

6 thoughts on “Do you need statistics to understand your data?

  1. paintingwithnumbers says:

    The point you make about qualitative data is a great one, its easy to get caught up in thinking that data is all about finding numerical significant or not and plotting graphs to show whats going on. Remember thought that if research finds that there is no significance that does not mean it is not successful as it has still provided proof that there is no effect etc. and it expands the knowledge in that research area.

  2. 1jessicakes says:

    “To me it feels like I’m actually getting to the point of the problem being investigated, rather than just using a load of numbers and sums to come out with one final figure which is supposed to be the answer to my problem!!” – This made me want to shout; “The meaning of life is 42!!!” I must agree with this argument.

    However, if we’re looking at our need for stats, stats can give us THE non-biased facts that are practically impossible to make when using qualitative research. The importance of stats is not to be overlooked, as every summation to create that final result plays such an important part! If we didnt understand our statistics, we could never rule out any outliers, or errors that may hide themselves within our data, and, as boring as quantative statistics may be, they do give us a much more accurate result*

    (Accuracy in Statistics, Cyril Davenport Hughes, Journal of the Royal Statistical Society. Series C (Applied Statistics) Vol. 3, No. 3 (Nov., 1954), pp. 146-149)

  3. rhiresearchskills says:

    I really enjoyed your blog!

    I thought you got your point across really well and have made me think more about it all.

    I do think we need statistics in psychology in both qualitative and quantitive data. In both we make comparisons, look for simularities in answers, corellate these answers. When comparing two treatment effects statistics help by comparing means, standard deviations and much more. By using statistics we can show this information in all sorts of ways like graphs.

  4. Spearmint123 says:

    I do agree that qualitative data is more interesting in the sense that it goes into so much more depth and detail. But I still think that statistics is still needed in all areas of psychology because like ‘rhiresearchskills’ says above, we still need the stats to show comparisons or irregularities in the results.For example if someone is researcher the effects of different treatments, one treatment being medication and the other treatment being a more verbal ‘talk about your feelings’ approach, there will be some form of statistical data produced at the end of the experiment so the results can be put together to be compared and reviewed to see which was most effective.

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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