Statistical methods in psychology

Updated on January 26, 2012

The ways to summarize and analyze the data

Some boring statistics should be used in order to come up with reliable research.

When conducting scientific research psychologists should find some way to summarize and analyze the data to see that the results of the research are trustworthy and reliable. Statistics is the solution here. We can categorize the statistical procedures in two groups:

1. descriptive statistics – which are used to summarize collected data
2. Inferential statistics – which are the collection of methods helping researchers to find out whether the results of their research is reliable.

I’ll try to provide brief introduction here of some commonly used statistics in psychology.

Descriptive statistics

Descriptive statistics is a simple scientific technique used to summarize sets of data. In order to measure average effects in which scientists are interested in, they use either mean or median. The mean is the arithmetic average and the median is the center score.

In most cases, researchers need to employ variability, which is the degree to which numbers in the set differ from one another or from their mean; and the common measure for variability is the standard deviation.

Another one statistic widely used in scientific studies is correlation coefficient, which measures strength and direction of the relationships between variables; it’s a strong instrument, because when correlation coefficient is strong enough, researchers can predict the value of one variable by knowing the other.

Inferential statistics

There is no research study in our reality, which will result in 100% reliability, because of constantly changing and unpredictable circumstances. That’s why inferential statistics is used when one’s interest is trustworthiness of research study. There always is some degree of variability in the set of data that can be attributed to chance in every research study. Think about implication of that: if the results in your study can vary due to chance, then how confident can you be when concluding final findings of your research?

Statistical significance is the way out: level of significance is the probability that a difference as great as or greater than that observed would occur by chance if, in the larger population, there were no difference between two means.

In short, a large observed effect, a large number of observations and a small degree of variability in scores within groups all increase statistical significance of the research study. The higher the statistical significance is, the more reliable and trustworthy becomes your research.