# Type of Statistics

##### Swamped with your writing assignments? Take the weight off your shoulder!

What is the difference between descriptive and inferential statistics?

There are two broad classes of statistics known as descriptive and inferential statistics. Descriptive statistics usually describe a sample. When summarizing and plotting graphs about a group selected for the study, descriptive statistics are applied. Thus, there is no degree of uncertainty when using the descriptive statistics since one only measures, analyzes and explains the specific sample used in the research work. Some of the measures used in descriptive statistics include central tendency, which entails mean, mode, and median. Another measure in descriptive statistics is dispersion which includes standard deviation, range frequency distribution. There is skewness which checks whether a distribution is symmetric or skewed.

On the other hand, inferential statistics are involved when observations about a sample are used to make conclusions regarding the larger population from which the sample was extracted. Inferential statistics is, thus, about generalizing findings made on a given sample to a larger population. The aspect of generalization brings up the issue of confidence, errors and uncertainty which are critical in inferential statistics. Inferential statistics entails hypothesis tests, regression analyses, and confidence intervals. The whole process of designing research, developing questionnaires, sampling, and conducting the data collection must, thus, be cautiously executed to ensure erroneous and invalid inferences are not made about the larger population.

What is the purpose of a descriptive analysis of the study population?

Descriptive analysis of a research population is meant to offer deep understanding of the broader study group in terms of its fundamental characteristics. For example, with descriptive analysis, a researcher is able to understand where most of the population values fall so that the best of fit line can be drawn. Descriptive analysis also helps a lot in identification of outliers so that datasets can then be cured. Outliers can easily affect the internal and external validity of a study. They must be known that explanations are only those caused by the independent variables, but not any other extraneous ones. Through descriptive analysis, dispersion of values within a given study population can also easily be established for better understanding. Descriptive analysis of the population prepares the researcher for inferential analysis. Inferential analysis is premised on descriptive analysis.

How do you compare descriptive and inferential statistics?

Similarities and differences can be drawn between descriptive and inferential statistics. Starting with differences, descriptive statistics give explanations about the characteristics of samples and populations. Descriptive statistics are thus used to get an understanding of facts about a dataset before use by a researcher. For effective research implementation, it is important that a researcher gets to understand the key features about both samples and population under study. Without this understanding, a researcher cannot be able to draw plausible findings and conclusions. In statistics, there are fundamental features about a dataset which researchers always target to know. The features are well-explained after conducting a descriptive analysis. Inferential statistics do not have this focus on the sample. On the contrary, inferential statistics focus on conclusions about the population. Further, inferential statistics do not seek to give information about the factual characteristics of the population, but inductive generalizations.

Therefore, whereas there is a high degree of certainty with the results from a descriptive analysis, uncertainty and a significant level of doubt always mars inferential statistics. It thus follows that inferential statistics, unlike descriptive analysis results, is based on assumptions. The assumptions applied in inferential analysis of populations must be explained to the audience. In most cases, the assumptions can be tested statistically with a view of ensuring their presence for valid findings and conclusions to be drawn. When the assumptions are absent, then findings made cannot be externally valid.

Unlike in the case of the inferential statistics, there are no errors and confidence levels needed in descriptive analysis because of the high level of certainty involved the results. Inferential statistics entails significance levels especially for hypothesis tests, degrees of freedom and the error margins. All these conditions that are set for a given inferential analysis show that a finding, and conclusion made about the larger population can be wrong, but close to the accurate position. Therefore, researchers cannot treat findings made about populations based on inferential analysis as accurate truths, but simply pointers of the true point.

With inferential statistical analysis, a researcher can effectively predict about the future albeit with some level of uncertainty. Descriptive analysis does not offer a researcher the ability to make any predictions about the future phenomena. It implies that inferential statistics is of more value and worth to policy makers and planners than descriptive statistics. However, results given about the future in inferential statistics are simply probabilities, whereas the findings of descriptive analysis are factual positions.

Similarities in the two classes of statistics include the fact that they both seek to inform the researcher and audience about a given population. In both cases, one can gain insights about a given population. The figure below shows the comparison between descriptive and inferential statistics.