Inferential Testing

Inferential Tests Psychology

In psychology research, researchers aim to identify if their results support their proposed hypothesis; raw data needs to be analysed to establish this. One form of analysis is called inferential testing.

Data analysis in research involves performing descriptive, statistical, and inferential tests. The tests are necessary to create summaries, determine the relationship between variables, and determine if the population's findings are generalisable.

If this is not the case, then the study should be revised as it has limited use in the real world.

Inferential Tests: Variables in Research and Chance

Before we get into the details of inferential testing, let's recap the different types of variables.

When conducting research, there will always be confounding factors to some degree. The main goal of research in psychology is to identify how the IV affects the DV. However, it is impossible for researchers to say with 100% confidence that their results are due to manipulation of the IV and not external factors or due to chance.

Types of Inferential Tests: Chance and Significance Levels

Chance is the theory that the results are not due to a fluke, e.g. the conditions during that day and a result of manipulating the IV. The probability and significance values of the research are measured to determine if the results are due to chance.

Inferential tests can be used to determine if a study's results are due to chance. Significance levels are a type of inferential test.

The value is best understood as a proportion. Let us look at an example that converts the p-value to a percentage.

The significance value of 0.05 is the recommended alpha value (another term for significance values) in psychology. The hypothesis should be rejected if a significance level above 0.05 is found. The reason is that the results are likely due to factors other than the IV. And the hypothesis should be accepted if the significance level is below 0.05.

The lower the significance level, the more likely the results are due to the intended variables being studied. Although it's next to impossible if the study was replicated on the entire population, similar results would likely be found. Therefore, the data can be considered suitable to generalise to the population.

We can never achieve a p-value indicating 100% confidence because, in research, we collect data from a sample of people who are most likely not representative of the entire population.

Inferential Tests in Research: Confidence Intervals

Confidence intervals are another form of inferential statistics that help researchers understand how representative their sample is of the general population.

Confidence intervals can guide how much the sample deviates from the population. If the data vastly differs, it is unlikely that it can be generalised to the population.

The variances in sampling confidence intervals and z-scores vary when different samples are used. This test differs from the previous inferential tests, z-scores, because it estimates whether the sampling procedure is representative of the population rather than the sampling distribution.

Inferential Statistics Tests: Hypothesis Testing

As mentioned earlier, errors sometimes occur when conducting experiments. These can be sampling errors, such as when the sample is not representative of the population or experimental errors.

Sampling and experimental errors can affect results and cause research to draw incorrect conclusions, such as type 1 and type 2 errors.

A type 1 error is when we reject the null hypothesis even though it is true (false positive), so the researcher believes their results are significant even though they are not.

On the other hand, an type 2 error is when the researcher falsely accepts the null hypothesis and rejects the alternative hypothesis when it is true (false negative).

Inferential Statistics Hypothesis Testing Examples

An example of an inferential test is the hypothesis test. The purpose of this test is to determine whether the results of the experiment are valid.

By estimating how likely the results are due to chance, we determine the validity of the results.

A null hypothesis must be stated to perform the test, and an appropriate statistical test should be chosen to perform the analysis.

In hypothesis testing, estimates of sampling error are considered to avoid errors in accepting/ rejecting the hypothesis and to reduce the likelihood of type 1 and type 2 errors.