Statistics without maths

SPSS step by step cover

‘SPSS step by step’ by Cole Davus

In his first blogpost, Cole Davis, author of SPSS step by step: Essentials for social and political science discussed his approach to quantifying the qualitative in statistics. Here he considers further approaches to statistics without maths.

In my previous blogpost, I demonstrated that quantification of qualitative data is not  misplaced. However, some objections to statistical testing may possibly have a rather different source, that of an aversion to mathematics in general. This may be camouflaged in some cases by objections to the usage of statistical tests, as already discussed, or it may reflect a clear desire to keep more than a barge pole away from any mathematical calculations.

It is my opinion that researchers do not have to learn mathematics in order to reach statistically sound conclusions. Just as most drivers do not need a working knowledge of the internal combustion engine to get across town, and graphic designers seem to get away without knowing anything about computer programming, then researchers can run statistical tests without needing to know or use mathematical formulae.

It is of course a little more difficult than that. The researcher does need to know what questions to ask and how to ask them, research design and test choice respectively. They also need to interpret the test results. With the help of a book which applies statistical theory clearly without requiring mathematical knowledge, both novice and experienced researchers should be able to use tests accurately and responsibly. There are of course some caveats.  It is recommended that they read the relevant material thoroughly; avoid entering all the data regardless of rationale; and eschew advanced tests until basic tests are mastered.

Many researchers do not often travel beyond the foothills of absolute numbers, averages and correlations. This does not mean that all advanced tests are to be avoided by newcomers to statistical testing.

Factor analysis, often the bane of postgraduate students’ lives, can be explained simply. If we have a wide range of variables, we may want to condense these to find a few meaningful underlying factors. This could mean that diverse questionnaire responses to a social issue may have a few common beliefs underlying them. The analysis will try to group together sets of responses to assist the researcher in making sense of the data. The interpretation of the outcomes still requires the researcher’s judgment, an art as much as a science.

The researcher may also want to create a predictive model. A set of relationships may be important in determining an outcome. Let us say that we have some reason to believe that the following (fictional) factors are related to schools’ examination success: teachers’ pay, the headteacher’s office size, the number of community policemen on the premises, the number of computers available and the number of library books. Each of these has a cost implication, so we want to find out if we can optimise spending by seeing if all are necessary or if one or two can be dropped without unduly affecting the results. Multiple regression is the tool of choice here. This type of study is now a regular feature of mainstream research and, as an understanding of multiple regression is a fairly logical progression from the use of correlations, it seems peculiar that most introductory books do not deal with this.

Not that experienced researchers should stop reading. As with previously discussed methods, an understanding of the issues under investigation will influence how they use multiple regression. In general, their work may be enriched by improving their knowledge of the relationship between tests and qualitative analysis.

Cole Davis is the author of SPSS step by step: Essentials for social and political science, published by The Policy Press on 13 February 2013.

0 Responses to “Statistics without maths”



  1. Leave a Comment

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




Enter your email address to follow this blog and receive notifications of new posts by email.

Twitter Updates

Archives

Creative Commons License

Republish our articles for free, online or in print.

The work on the Policy Press blog is licensed under a Creative Commons licence.