Statistics in Management: Descriptive vs.

Statistics in Management: Descriptive vs. Inferential Statistics

The use of descriptive vs. inferential statistics in organizations provides decision makers, managers and leaders with the necessary insights to compete more effectively in an increasingly challenging global economic climate. The intent of this essay is to define which conditions are optimal for the use of each. Descriptive statistics by definition are more adept at the consolidating of data and its summarization (Spatz, 2008). Inferential statistics however are meant to be representative of a broader population and are developed to be statistically sound (van den Besselaar, 2003). The use of each of these types of statistics varies significantly within organizations, and has completely different interpretations when used. This essay examines how each are used to their optimal value.

Best Practices in Descriptive Statistics

There are several functional areas within organizations that rely heavily on descriptive statistics. These include accounting, business planning and analysis, financial planning, marketing, sales, product management, quality and production. Each of these functional areas are often evaluated on scorecards and benchmarks-based entirely on descriptive statistics of their activity over time (Ainslie, Leyland, 1992). Best practices in descriptive statistics for example in marketing centers on the need to accurately and succinctly summarize customer feedback about existing marketing strategies, experiences with customer service centers, and the prices paid for products as well. Descriptive statistics is an indispensible tool for evaluating which strategies are best used for retaining and growing customer loyalty as well (Ainslie, Leyland, 1992). In the area of production, descriptive statistics are very useful for evaluating the effectiveness of production techniques, systems and routing of specific products over the shop floor. This is exceptionally valuable for getting greater performance and production from less space, as lean manufacturing techniques rely on descriptive statistics for insights into how to continually improve. As these examples within organizations indicate, descriptive statistics are best for creating a synopsis or summary of a given set of variables that have a major impact on the organization. From customers to suppliers and production processes, descriptive statistics are invaluable for gaining insights into how to improve an organization over time.

Inferential Statistics and the Defining of Strategies in Organizations

The conditions for applying inferential statistics in an organization are when the data has been statistically and reliably collected to reflect a broader population of users. Inferential statistics are best applied when a valid methodology and sampling frame have been defined and the statistics can be reliably extrapolated to a broader population. This type of measurement is best used when the data has also been captured at the ordinal or ratio level as the orthogonality of the data set is reliable (Marshall, Ruiz, Bredillet, 2008). Extrapolating statistics to a broader population is also dependent on the approach of randomization used. When a solid methodology, sampling frame and approach to randomization have all been defined, inferential data is often used in organizations to define company-wide strategic initiatives. These include the decision to open more retail stores for a clothing or sporting goods retailer for example based on demographic data.

Inferential statistics are often used for experimental analysis where the statistical significance of relationships within the data is analyzed is highly useful for gaining insights into customer preferences and requirements (Ainslie, Leyland, 1992). Using inferential statistics to project the cause-and-effect relationships of marketing and sales programs has proven to have a high Return on Investment (ROI) as well (Marshall, Ruiz, Bredillet, 2008).