Formulate clear research questions that describe what you want to learn from your statistical analysis. You may want to learn, for example, the average grades earned by students in a classroom or the impact of additional education and training on the average incomes of computer programmers. The question you try to answer will help determine the type of analysis you conduct on your statistical data. The question about average grades, for example, may require only a set of descriptive statistics, such as means. The question about education and training's effect on earnings, however, requires more complex analysis.
Collect and enter your data. Spreadsheet programs such as Excel, Numbers and Open Office are excellent for entering data. While entering data into a spreadsheet or database, you may need to alter the format to meet your analytical needs. For example, if your data comes from a questionnaire or a set of interviews, you will have to assign numeric codes to the responses so that they can be analyzed with statistical techniques.
Examine your data visually before embarking on more complex analyses. This is an important step; yet even trained statisticians and data analysts fail to take the time to examine the data first before analyzing it. A useful method for visualizing the data is to produce a graph that summarizes the information. Excel and statistical software programs have the capability to do this. The type of data you are analyzing will help determine the most appropriate type of graph. Pie charts, for example, are an excellent way of summarizing financial data, while bar charts can visually represent survey responses.
Conduct your analysis, using a spreadsheet or statistical software program. Study the results closely, looking for patterns in the results that provide answers to your research questions.
Report the findings of your statistical analysis using charts and graphs, where possible, to present your results. Visual displays have greater impact among the intended audience of your analysis than blocks of written text. Use the text of your report to highlight the key findings of your analysis.