# Statistical Communication for Rating, Ranking, Sorting Data

### Choosing an Effective Communication Method for Statistical Studies

The decision of an optimal communication method occurs when a researcher has determined the best data collection approach.  In the current business environment, technology has produced extensive changes to the communication methods commonly applied in business research. In the past, the paper and pencil survey as a common approach, by today nearly all surveying takes place in a digital format. In order for the communication approach to be successful, it is important for the researcher to minimize errors from both interviewers and participants. Interviewer error can occur when the interviewer has failed to execute the process correctly or has exhibited a bias that influences participant responses. Participant error can occur if the individual is not motivated to focus or does not fully understand their role in the study. An effective researcher should be able to recognize these errors and take action accordingly to mitigate their impact on the study.

### The Four Main Scale Classifications

Measurement is the process by which elements within a dataset are compared to one another. There are four main classifications of scales, which include the nominal, ordinal, interval, and ratio. Data on the nominal scale only contains arbitrary labels and non-numeric values. Nominal scales are most useful when conducting qualitative research. The ordinal scale contains values that are sequenced a specific order, but they are not separated by equal magnitudes. The interval scale contains values like the ordinal scale, but they are separated by the same order of magnitude, such as degrees of temperature in Celsius.  Lastly, the ratio scale contains values that have equal intervals and an absolute zero origin value, such as degrees of temperature in Kelvin.

### The Differences between Rating, Ranking, Sorting

Rating, ranking, sorting, and preference scales excel in different research applications. The rating scale is commonly used to understand the total overall perception of a group, such as “On a scale of 1 to 5, how much did you like XYZ film?”. This can be used to understand how well a product or service fares compared to competitors. A ranking scale requires respondents to order items based on their preference. This type of scale is used to discover consumer favorites and brand leaders. A sorting scale is more complex and requires participants to group items based on specified characteristics. The preference scale asks multiple similar questions, then provides a cumulative score based on the results. This allows multiple related topics to be combined to understand a topic from a comprehensive perspective.

### Deciphering the Information Obtained

Information received via research questioning is typically classified into three subtypes, which include target questions, classification question, and administrative questions. Administrative questions are used to obtain general background information about the participant and do not gather information that is specific to the research objectives. This step is critical because it builds a demographic profile that is central to many studies. Classification questions are used to obtain qualitative information that can provide insight into the participant’s responses. This step is useful for understanding what drives a participant to respond a certain way. Finally, target questions are designed to answer the main objective of the research problem.  The target questions will address the specific data collection model required for analysis in the study.