UNIVARIATE ANALYSIS

Uni variate analysis is a single-variable analysis.  In a questionnaire-based marketing research project, each question often represents a variable under study.    Simple tabulation involving single variable constitutes uni variate analysis. In most marketing research applications, a survey of some sort is the method used, whether it is conducted through mail, a personal interview, over the phone or more recently, on the Internet.  There are however, other classes of study available, one of which is observation.  The other widely used class of study is known as experimentations.  Just like in a laboratory, we manipulate certain variables (usually marketing related ones in marketing research), and observe changes in other variables (for example: sales, or consumer behavior, or attitude) 

 

 WHAT UNI VARIATE ANALYSIS DOES?

•   Explores each variable separately in a data and considers the range of values, and the central tendency of the values.
•  States the pattern of responses to the variable.
• Asserts each variable on its own. Descriptive statistics asserts and summarize data. Uni variate descriptive statistics indicates individual variables.

How to analyze one Variable 

• Raw Data: Obtain a printout of the raw data for all the variables under study. Raw data is similar to a matrix, with the variable names heading, the columns, and the data for each case or record displayed across the rows. Raw data is difficult to absorb, especially with large number of cases. Univariate descriptive statistics may summarize large amount of numerical data and expose patterns in the raw data. To present the data in a more organized way, start with univariate descriptive statistics for each variable. 
•  Frequency Distribution: Obtain a frequency distribution of the data for the variable accomplished by identifying the highest and lowest values of the variable and then putting all the values of the variable in order from lowest to highest.  Then count the number of appearance of each value of the variable. This is a frequency count with which each value occurs in the data. 
• Grouped Data: Decide on whether the data should be grouped into classes. One way to construct groups is to have equal class intervals. Another way to construct groups is to have about equal numbers of observations in each group. The class intervals must be both mutually exclusive and exhaustive. 
•  Cumulative Distributions: Cumulative frequency distributions consists a third column in the table 
• Percentage Distributions: Frequencies can also be presented in the form of percentage distributions and cumulative percentages. 
• Graphing the Single Variable: Graphing is a way of visually presenting the information. Several individuals can grasp the information presented in a graph better than in a text. The aim of graphing is to:
• present the data
• summarize the data
• enhance textual descriptions
• describe and explore the data
• make comparisons easy
• avoid distortion
• provoke thought about the data  

TECHNIQUES OF UNI-VARIATE ANALYSIS

Bar Graphs: Used to display the frequency distributions for variables nominal and ordinal levels measure. Bar graphs employ the same width for all the bars on the graph, and space between the bars.   
Histogram: used for interval and ratio level variables. In a histogram, the width of the bar is important, since it is covers the total area under the bar that represents the proportion of the phenomenon accounted for by each category. The bars present the relationship of one group or class of the variable to the other(s).

 Frequency Polygon: displays information for an interval or ratio level variable which displays the area under the curve that the values of the variable represent. Also used to show time series or the changes in rates over time. A cumulative frequency polygon is used to display the cumulative distribution of values for a variable.

Pie Chart: shows the relationships between classes or categories of a variable is in a pie or circle chart. In a pie chart, each “slice” represents the proportion of the total phenomenon that is due to each of the classes or groups.