Statistical analysis is a quantitative method to find probabilities between sets or results of data. This data can come from the natural or social sciences. Statistical analysis helps elaborate on trends or patterns found within the research of a topic. For example, if a medical doctor needed to test the probable effectiveness of a drug, she would utilize statistics to see if the drug worked a certain number of times for a certain population. Then, she would analyze the results to see if the drug results have a relative or probable accuracy.

See if the null hypothesis has been disproved. The null hypothesis is, basically, the opposite answer to the researcher's question. If the analysis does not clearly disprove the null hypothesis, then the researcher's statistical analysis is questionable.

Investigate the quality of the data. A researcher needs to show where he is getting his data. Critics can evaluate whether the statistical analysis utilizes incorrectly measured data in order to cater the data to the research.

Utilize an evaluative means test such as the ANOVA test. An ANOVA, or analysis of variance, test is an evaluation tool that makes sure that averages exist within each variable test group. If not, then the sample sizes in the statistical analysis may be incorrect.

Set up a regression. A regression is a general statistical tool that sees how variables are connected. For example, if a researcher states that x causes y, a regression would calculate whether x always led to y in different scenarios. If the regression shows the researcher's answer to be false at least once, then the statistical analysis is suspect.

Analyze the qualitative aspect of the analysis. All researchers summarize what the data means for their field, whether it is biology or public policy. You can assess the entire scope of the researcher's study and conclude a different take on the researcher's analysis.

### Tip

There are different types of means tests within the statistical world. A similar tool to the ANOVA test is the t-test. The t-test compares sample groups to see if there are averages within the continuous variables and nominal variables of sample. Continuous variables are changeable variables such as height, while nominal variables are consistent variables such as blood type. Such means tests are popular in the social and natural sciences.