2022 Introduction to Statistics in Research Mitchell 2nd ed

I N T R O T O R E S E A R C H : D A T A V I S U A L I Z A T I O N & C O M M O N S T A T T E S T S

Chapter 3

What test do I use?

The answer to “What test do I use?” is that depends .

Researchers are good at collecting data, but often less savvy at turning it into actionable insights or theory building. First, let's discuss several types of statistics (some of the language is borrowed from business)

1) Descriptive statistics – describes and summarizes the data. Through visuals, you can describe a trend or specific features. Tools used to visualize descriptive statistics include bar charts, pie charts, histograms, scatter graph, boxplots, etc. Addresses the question, “What happened?” Numerical descriptive tools allow you to summarize data like average, mean, mode, and minimum and maximum. As you learned earlier, there are two types of statistics used to describe data:

a. The measure of central tendency (mean, median, or mode)

b. The measure of spread through range, quartiles, variation, and standard deviation

2) Inferential statistics – the word inference is a deduction or conclusion. Inferential statistics is a technique that allows us to draw inferences or conclusions about the population from the sample. The researcher accepts that sampling incurs sampling errors, and the researcher should not expect perfect representation of the population. The tools used most often include hypothesis test,

3) ANOVA Inferential statistics is a big part of Six Sigma. Shown in the illustration here we see the empirical rule of standard deviation at one (68.2%), two (95.4%) and 3 (99.7%). Six sigma takes it much further. They work with up to 99.9999998% data points which is ± six standard

Table 105: A normal distribution showing the emperical rule

deviations from the mean . According to Master of Project Academy, “Six Sigma is about creating a culture that demands perfection and that gives employees the tools to enable them to pinpoint performance gaps and make the necessary improvements using data-driven problem-solving methods” ( para. 10). Academic researchers are less likely to work with data that requires this level of focus on acceptable levels, but practitioners might be required to work with the six sigma model. For example, the quality of a new drug would require this model for quality control. The

110