Data Presentation Methods
Chapter 3 in Core Text
Key points in this chapter :
Ñ Basic Terminology
Ñ Real and Money Terms
Ñ Data Presentation Methods
Ñ Kinds of Diagrams
Ñ Statistical Tools of Decision Making
Ñ Collection of Data
Inorder to make various decisions effectively, it is essential that managers have access to a wide range of data. However, for data to be used effectively it should be summarized, classified and presented appropriately. There are various tools and techniques of data presentation that help managers in this regard.
Variables: the term variables refers to whatever characteristic we are investigating or analyzing. Examples may include, company profits, salaries, number of employees etc.
Discrete: a discrete variable is one, which can only take certain fixed numerical values. For example, the number of cars sold by Ford, the number of employees in a company etc. These can only expressed in whole numbers and not as fractions or sub-divided units.
Continuous: a continuous variable is one, which in principle can take any numerical value. The length of a steel sheet for example can be expressed to any degree of accuracy - centimeters, millimeters, hundredths of a millimeter and so on. In the same way, we can measure weight, exchange rates, etc to a high degree of accuracy.
Attributes: an attribute variable is one that is not normally expressed in numerical terms. The level of education achieved by students is not a variable that can be expressed sensibly in numerical terms. Other examples may include people's moods, lifestyles, opinion towards a particular concept or idea etc. Attributes are qualitative in nature and hence require a more detailed examination for analysis.
Inorder to make an effective comparison between two variables, spread over a period of time, it is highly essential to consider the real changes and not merely the money changes. For example, if we compare the total sales of a company in 1990 and 2000, the sales may have increased by 20 %. However, if during the same period the selling price of our goods has increased by 25 %, then there is no real growth in the sales. It is merely the total sales amount that has increased and not the number of units sold. Thus to make an effective comparison over a period of years, it is advisable to compare using a base year. In the above example, the sales figure of 2000 should be calculated at the price levels of 1990 to make a comparison in real terms.
With the help of various diagrams, complex facts may be shown in an easy way.
The facts and figures become more attractive.
Various facts may be compared more easily
Some of the most frequently used graphs and diagrams are:
Bar graphs – subdivided bar graphs, multiple bar graphs, percentage bar diagrams
Measures of Central Tendency: central tendency, which is also called average, presents the vast data in a precise form. An average is the single value within the range of data used to present all of the values in the series. Following are the types of averages:
1. Arithmetic mean or simple mean
Measures of Variability: the degree to which numerical data tend to spread about on average value is called the variation or dispersion of the data. Following are the methods of measurement of dispersion:
1. Limits Method:
· Inter Quartile Range
· Semi Inter Quartile Range
2. Deviation Method
· Mean Deviation
· Standard Deviation
· Quartile Deviation
3. Graphic Method – Lorenz Curve
Measures of Correlation: the process by which the relationship between two or more variables is found, is called correlation. Correlation may be:
Positive: when the change in the values of two series are in the same direction, there will be positive correlation. For example, if prices are increasing, supply is also increasing.
Negative: when the changes in the values of two series are in opposite direction, there will be negative correlation. For example, if the prices are increasing, demand is decreasing.
Following are the methods of measurement of correlation:
1. Scatter Diagram
2. Graphic Method
3. Karl Pearson’s Method
Data may be classified into two broad categories:
Primary Data: primary data are those data that are collected for the first time by the investigator for the purpose of present investigation. For example, if the marks of the students of a class are to be collected and they are collected by the investigator himself, by getting information from the students themselves, it will be known as primary data. Methods of collecting primary data include direct personal investigation, indirect oral investigation, surveys, focus groups, interviews, web-based reply forms etc.
Secondary Data: when some data were collected by someone else and is being taken by the investigator for his investigation also, it is called secondary data. For example, if the marks of the students of a class are collected from the record of a school, it will be known as secondary data. Methods of collecting secondary data include published data like international publications of UNO, government publications, newspapers and magazine publications, publications of business institutions, individual publications etc.
External Validity is the degree to which the conclusion in your study would hold good for other persons in other places and at other times
The theoretical population is the population on whom we want to generalize our research findings.
The study population is the population we can get access to.
The sampling frame is the source through which we can get access to our respondents.
The sample is the actual set of people who are included in our research.
Some frequently used sampling techniques are as follows:
1. Simple random sampling – every item has an equal chance of being represented in the sample
2. Stratified random sampling – first divide population into various strata (divisions) and then select randomly from each strata.
3. Systematic random sampling – every nth item is included in the sample.
4. Cluster random sampling – first divide the population into various clusters (usually geographical) and then randomly select clusters, then measure all units within the selected clusters.
5. Multi-stage sampling – using different mix of sampling techniques at different levels
1. Convenience sampling – the researcher collects information from only those people to whom he has an easy and convenient access to.
2. Purposive sampling – the researcher may first judge a person and asks only those to participate whom he ‘thinks’ fall into a group.
3. Quota sampling – the sample is taken on the same quota as the percentage distribution of variables in the population
4. Snowball sampling – after collecting information from one respondent, the researcher may ask the respondent to suggest/recommend some other people or friends who would be willing to participate in the research.
Decision-Making Academic Homepage