Dissertation Statistics and Thesis Statistics


Posted December 30, 2008

There are a dizzying number of statistical tests out there and knowing where to start can be a real problem. You probably haven't taken a statistics class in years and now you are being called upon to recall statistical tests and details that you believe even a statistician would not know. Fear not! Read on and click through to get all the help you need.

What are dissertation statistics and thesis statistics?

Dissertation statistics and thesis statistics are the statistics used in a dissertation or thesis… just kidding – okay seriously. You have spent all kinds of time on the internet and the library completing this epic task only to hit the wall of………statistics.

There are all kinds of statistics you could use for your Master's thesis, Master's dissertation, Ph.D. thesis, and Ph.D. dissertation. These days, it is assumed and maybe required that you use multivariate statistics of some kind. The days of simple bivariate correlations and t-tests seem to be gone forever – depending on the area of expertise. Given the increased level of familiarity with the tests like multiple regression, logistic regression, n-way ANOVA, mixed ANOVA, ANCOVA, MANOVA, MANCOVA, and the like, institutions and committee members' expectations for you thesis or dissertation are much higher. The individual tests, their benefits, and their uses are described in some more detail here.

What statistical analysis should I use for my thesis or dissertation?

The statistical analysis for your thesis or dissertation should be appropriate for what you are researching and should fit with your needs and capabilities. I know, that's not saying much, but it's important that you're comfortable with the statistical analysis you will be conducting. An experienced dissertation consultant will help you tremendously with this. You will find a great one here.

I am going to go briefly cover why multivariate analysis are so popular with institutions and committees, but before your head explodes, keep in mind that seeking professional help will simplify this greatly by saving you the hours of researching the tests and revising if – God forbid – the tests are run incorrectly. In the end, an experienced dissertation consultant is worth his weight in gold.

What are multivariate statistics and why should I use them for my dissertation or thesis?

The main thing multivariate statistics do is protect your alpha. Something happens when you conduct multiple tests on the same dataset, e.g. t-tests and correlations: you run the risk of the statistical tests being falsely significant simply because multiple tests were conducted on the dataset. In other words, your statistical tests may show significance even though they aren't.

There are corrections for this like the Bonferroni correction, however, a researcher can avoid the stringent thresholds that follow with implementation of the Bonferroni correction by simply conducting multivariate statistics for their dissertation or thesis. Instead of conducting multiple tests on the same dataset, maybe only one or two tests are conducted. Get help using multivariate statistics with your Master's thesis, Master's dissertation, Ph.D. thesis, or Ph.D. dissertation.


Pin It on Pinterest

Shares
Share This