ellies1mpson


“Do you need statistics to understand your data?”
October 7, 2011, 5:10 pm
Filed under: Uncategorized

Statistics are not necessary for qualitative data.  Qualitative data can give rich, vivid insight into phenomena, answering questions of why or how, but as they are not numerical, statistics are not used for analysis.

For quantitative data statistics are necessary to be able to effectively conduct research and for making sense of raw individual data.  Statistics are a powerful tool for condensing and analysing large quantities of data.   They are standardised techniques that are recognisable and understood throughout the scientific community.  They enable information to be presented and interpreted in both an accurate and informative way, so we can make informed decisions based on the data collected.

Statistics can tell us whether one variable has an effect on another, what that effect is and the size of that effect.  They show the distribution of the data enabling us to identify patterns in the data such as averages showing the norms.  They show the similarities and differences in data and what that these are and also relationships, correlations between variables.

With inductive reasoning, to study a relatively small sample and to draw a conclusion appropriate for people in general, one can never study a sample and expect conclusions to hold true for the entire population with absolute certainty.

With deductive reasoning, going from the general to the particular, uses the logic of mathematics so is therefore absolutely certain. General principles logically conclude to more specific relationships.

Inferential statistical procedures allow researchers to estimate, make probability predictions assessing the likelihood that the collected data occurred by chance factors, indicated by the level of significance.  The likelihood of the results occurring by chance must be very low for them to be deemed significant. Inferential statistics enable researchers to infer that the results obtained in data from a sample would also occur in the larger target population from which the sample was drawn, therefore can draw conclusions about a larger population from which samples were collected.

Statistics are a bridge between the inductive uncertainty of science and the deductive certainty of mathematics, and are a necessity to unravelling and understanding quantitative data.

Having said all that be aware statistics can be manipulated!

Advertisements

13 Comments so far
Leave a comment

I have to disagree with you on this matter, as Statistics do not tell you if one variable has an effect on another, it merely tells you if a variable changes when another changes. To illustrate this I will give an example; If a sample of children is used to find out if violence on television causes violent behaviour in children, the children would be observed in terms of how violent they are, and what kind of television programmes they watch.
If the data states that children who watch the most violence on television show the most violent behaviour, is it fair to say that violent television programmes causes children to be violent? Statistics does not allow us to draw that kind of conclusion. It could be possible to say that violent children are more likely to want to watch violent television programmes.
Statistics do not give the cause and effect answers, it merely gives us a set of data, for which we draw our own conclusions.

Comment by psud78

I feel your argument is merely nit picking. Obviously statistics are just figures, but the analysis of this data is what gives us the answers to our questions. As psychologists I automatically assumed you’d understand I meant processed data, but my fault for not making this explicitly clear.

Comment by ellies1mpson

I don’t quite agree with your statement that statisitics is not needed for qualatitive research, yes qualatitve is richer and deeper, but there are methods to transfer findings into numerical data that can be analysed statistically. although, not as important as the need for statisitics in quantative research, statistics can still give further insight into findings and results.

I agree that statistics are important in quantative research, but as I mentioned in my own blog, believe that a deeper understanding is required, which is often lost through statistics, and up to us as individuals to recognise when statisitcs is sometimes just not enough. We can establish cause and effect, but why and how are important questions often lost in quantative research, which is when further research is required.

Comment by prpfe76

Many of the comments made here are about analysing qualitative data by encoding into something that can be tested using objective statisical procedures. The point of qualitative data is to be subjective, to give individual meanings. Perhaps this explains why such findings get lost in research.

Comment by ellies1mpson

I disagree that stats is not necessary for qualitative data, yes it is harder to analyse through stats because it is normally observational studies on behavior. But by coding based on themes in your data for example, can convert the qualitative data in to quantitative. Making it easier for you to analyse your findings and preform statistical tests.

Comment by itsstats3453

Many of the comments here are about analysing qualitative data by encoding into something that can be tested using objective statisical procedures. The point of qualitative data is to be subjective, to give individual meanings. Perhaps this explains why such findings get lost in research.

Comment by ellies1mpson

Your blog had some really informative points, which I agreed with. I especially agreed with the point you made about the statistics being used to show the patterns in the data, why are these patterns so important to us as psychologists? Do they improve your knowledge of the results or around the subject? I also agreed with your point where you said about statistics being used as a means to display the data correctly, I also agree that it is very important to display the data correctly however do you think that even though the data is displayed correctly that we have seen all the different avenues the data has to offer? Do you think statistics shows us the fuller picture of results and the different reasons why we gained those results? I really enjoyed your blog and I agree with your last point of statistics being essential for quantitative data but do you think that qualitative data can be used in statistics and is it just as important as quantitative data?

Comment by standarderrorofskewness

Many of the comments made here are about analysing qualitative data by encoding into something that can be tested using objective statisical procedures. The point of qualitative data is to be subjective, tabout individual meanings. Perhaps this explains why such findings get lost in research.

Comment by ellies1mpson

[…] https://ellies1mpson.wordpress.com/2011/10/07/%E2%80%9Cdo-you-need-statistics-to-understand-your-data… Share this:TwitterFacebookLike this:LikeBe the first to like this post. […]

Pingback by Homework For My TA Week 3 « Standard Error of Skewness

Similar to the other comments i disagree with the statement that statistics are not necessary for qualitative data. This is because by not using statistics to analyze qualitative data it leaves the data to be analyzed in a subjective way so it’s open to interpretation therefore reducing the reliability of the research as different researchers may analyze the same behaviours differently. Whereas if a researcher uses coding to convert qualitiative data into numerical data it will mean that statistics can be used to analyze the data objectively so that nobody can disagree with the findings. So if Psychology is to be considered a Science it’s essential to analyze all data scientifically which must involve using statistics to analyze ALL data objectively

Comment by liamjw91

Many of the comments made here are about analysing qualitative data by encoding into something that can be tested using objective statisical procedures. The point of qualitative data is to be subjective, to giving individual meanings. Perhaps this explains why such findings get lost in research.

Comment by ellies1mpson

Many of the comments made here are about analysing qualitative data by encoding into something that can be tested using objective statisical procedures. The point of qualitative data is to be subjective, to give individual meanings. Perhaps this explains why such findings get lost in research.

Comment by ellies1mpson




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



%d bloggers like this: