A data analyst transforms the data found inside of an organization into new information critical to business decisions. Her task may be as simple as gathering a list of customers with a certain credit rating, or she may perform a complex time series analysis to predict next year’s sales trends. Anyone with a good background in math, critical thinking skills and attention to detail can excel in this career. Median salaries are around $74,000 per year.
Before an analyst can start her work, she must collect the data to be studied. Data may come from a corporate database, a survey, supplied from government agencies or downloaded from Internet sources. Analysts may design and administer surveys, craft the questions, notify customers of the survey, then gather and analyze the results.
Once the data has been collected, the analyst often must filter and "clean" the data to remove errors and ensure consistency. Abbreviations found inside names and addresses may not match, data values may have been keyed incorrectly or fields may be missing. Analysts often need some programming skills to correct errors or may use programs like Microsoft Access to repair data.
Lists and Summaries
Many of the reports an analyst creates simply filter, sort and list data, adding groupings and subtotals. Tools like Microsoft Access or Crystal Reports allow the analyst to design these reports by dragging and dropping data elements and titles onto a template. When executives need data, they often describe their needs in general terms. Then, it is the analyst’s job to determine the source of the data and design the appropriate list or summary.
Graphs and Charts
In addition to lists and summaries, data often is presented as charts and graphs. Using tools like Excel or other presentation graphics packages, the analyst filters and groups the data, then selects a format that effectively presents the information.
When a list or graph does not meet the executive’s needs, the analyst will dig deeper into the data, using statistical techniques like correlation, regression or time series analysis to then write a report summarizing the findings in a language the executive understands. Tools like SAS or SPSS allow the analyst to create a variety of statistical models from a dataset, extracting averages, standard deviations and other statistical summaries. The analyst must know which statistical techniques apply to the given problem, and then translate these results into information useful to the executive.
Statistical analysis must be translated into meaningful information, so much of an analyst’s time is spent writing documents and reports. Findings may be published in research journals or may be described in executive summaries, so an analyst must have strong writing and communication skills. All of the math and analysis will be meaningless if the analyst cannot translate the knowledge into words the executive understands.