Summary of Descriptive and Inferential Statistics
When one thinks of data, usually rows and rows of data comes to mind either in the form of a spreadsheet or using a database tool. It is the analysis of this data, summarizing, tabulating, cross-referencing, and discerning patterns that makes the data valuable for researchers. For it is this, that tells the story. Descriptive statistics is the term given to the analysis which finds patterns in data. Descriptive statistics is usually achieved through the use of summary tools …show more content…
Descriptive statistics usually involve measures of central tendency such as mean, median, mode, and measures of dispersion such as variance and standard deviation. Descriptive statistics refers to the underlying data and does not draw conclusions about the data being represented. The strengths are you can clarify large volumes of data with no uncertainties. The weakness is there are no generalizations about the data and the results are not 100% accurate.
Inferential statistics refers to a sampling of data, and does not refer to a whole data set. Inferential statistics is useful when analyzing very large datasets where an estimation is would give a level of confidence. The strengths of inferential statistics allow the researcher to make generalizations about a dataset, or in most cases. The main weakness is the entire dataset is not fully measured, therefore a researcher cannot be completely sure about the results. The second weakness is inferential statistics require the researcher to be able to make an educated guesses to run the inferential tests. This introduces the possibility of bias into the …show more content…
For example, the probability of any given heard rate can be shown by a histogram. Ian Kestin wrote an article telling the basic principles of statistics in medicine. The topics include types of data, descriptive statistics, using the mean, median, mode, percentiles, the normal distribution, confidence intervals, and the standard error of the mean, hypothesis testing and the choice of statistical tests, type 1 and 2 errors, contingency tables, correlation and regression, and meta-analysis. When dealing with topics where exact data is needed, descriptive statistics is the best