A quantitative study will involve at least some form of statistics – from basic descriptive statistics to more complex analyses. The choice of statistical techniques is a very important one and should already be made at the proposal stage.
The research questions and hypotheses of your study will determine the statistical analyses needed. It is very important that you already know at the proposal stage which techniques you are going to use. If you need assistance in determining which techniques to use, we can assist you.
The choice of technique depends on the level of measurement of your variables. There are essentially two types of variables – categorical and continuous. Categorical variables are exactly what the name says – it provides data in categories. Typical examples are gender, age categories, or yes/ no questions. It can therefore take on a limited number of values.
There are two types:
Nominal data is data where there is no inherent order to the categories. “Nominal” comes from “name” – categories are therefore simply named. For example, for gender, we may code male a “1” and female a “2”, but that does not mean that female is “more” than male in any way! (Well, maybe……. 😊) We could just as well have done it the other way around. Other examples are language, job category or nationality.
Ordinal data is categorical data where there is a meaningful order of categories.
We know that a person who ticks “3”is older than a person who ticks “2”, but we cannot determine how much older! Therefore there is order in the categories – it is not simply names.
One example is age categories:
Continuous data where the variable can take on any number of values. Height or weight are examples of continuous data – it can take on a very wide range of values, and can also take on decimal values. Something like the number of chairs in a venue or the number of people can also take on a wide range of values, but cannot take on decimal values. They are still examples of continuous variables.
Choosing statistical technique
So, the first step in choosing the correct statistical technique is to identify the level of measurement of your variables. From a statistical viewpoint, it is always better to go for continuous data. As soon as you categorize variables which are inherently continuous (such as age or years of service) you loose information. If we know that someone’s age is 35, or 40, we know a lot more than just knowing these 2 people fall in the category of 35-40 years.
The next step is to identify your independent variable dependents variable. The independent variable has an influence on the dependent variable. For example: You design a study to test whether changes in room temperature have an effect on math test scores. Your independent variable is the temperature of the room. Your dependent variable is math test scores. Similarly, if you want to know whether males and females differ in math performance, gender is your independent variable and math performance your dependent variable.
Don’t be confused by sources which suggest that the independent variable is the “cause” of the variable that you “manipulate” – this is applicable to experimental research, which is seldom performed.
Once you have identified the level of measurement of your variables, you need to decide whether the particular relationship you want to measure is concerned with a difference between groups, or the strength of the relationship. Knowing what you are interested in, guides you in making the correct decisions in terms of statistical analysis. If you need more assistance click the button below.