What is statistics?
Statistics is a scientific approach that deals with a range of procedures for collecting, gathering, organizing, summarizing, presenting, and analyzing data and making inferences to aid in finding solutions to problems (Jani, 2014).
Contrast between qualitative and quantitative data
Quantitative data is that which can be measured and reported numerically (Brown & Saunders, 2008). For instance, the marks scored by students in a test, the GDP of countries, among other examples. Quantitative data can be either discrete or continues. Continuous data can take any number hence includes decimal points. For example, the distance between two towns, the height of students in a class, among others. Discrete data can only take whole numbers thus cannot be measured in decimals. An example is the number of students in a class.
Qualitative data, on the other hand, is that which cannot be measured numerically. Only the frequencies of a given category can be given numerically. For instance, the gender of students in a class is a qualitative data. A student is either a male or female. However, we can determine the number of students in each category; the number of males and that of females.
Tables and charts used to represent data
After collecting and summarizing data, it is described in a manner that is easily understandable to the potential users. The description is carried out by representing both qualitative and quantitative data using charts and tables. Charts and tables used in rendering data include pie charts, bar graphs, frequency distribution tables, histograms, among others (Jani, 2014).
A pie chart is a circle showing the composition of a population. It shows the percentages or portion of a population that is contributed by each component. Pie charts are used in representing categorical data, whether qualitative or quantitative. For instance, it can be used to describe the results of a survey of the popularity of 4 presidential candidates. The total number of respondents surveyed is the population. From the pie chart, it is easier to determine which presidential candidate has the highest popularity without carrying out any detailed analysis. Pie charts are useful in representing data, however, when the data has numerous categories, the use of pie charts will be limited.
Bar charts use rectangles with equal width to show the differences between the classes of data. The group with the highest value or number will have the longest bar while that with the least value will have the shortest bar. From the bar chart, one can easily identify the category with the highest value. It is useful in representing qualitative data. For instance, a business can conduct a survey by interviewing 2,000 customers on the attributes they value most in a product. The company can give four alternatives such as low price, high quality, ease of use and be locally produced. The number of customers who chose each of the four attributes is recorded and represented by a bar with the height of each bar showing the number of clients who preferred the quality.
The Frequency distribution table is mostly used in representing quantitative data. It lists the categories of data with the corresponding rates. For instance, this kind of distribution is used in describing the results of a Mathematics test for a class with 200 students. It is not possible to represent each student as a category. A frequency distribution table helps in categorizing the data by creating classes/intervals. For example, the first group could be students who scored between 0 and 20 marks. It makes it easier to determine which group has the most number of students. It also helps in representing the data using a histogram as well as other tables and charts.
Levels of data measurement
The nominal level of measurement: this level of measurement only categorizes data. Numbers, words and letters are used to categorize data. There is no ordering or ranking of data in this level. For instance, two players can have jersey numbers 15 and 30. The difference between 15 and 30 is not significant.
The ordinal level of measurement: Attributes can be ranked on a scale. For instance, the marks scored by students can be ranked from the highest to lowest. Some qualitative data can also be ranked. For instance, strongly agree, agree, disagree and strongly disagree.
Interval level of measurement: This level classifies and orders attributes and also specifies that the interval between is constant. For instance, the difference between 83 and 85 is equal to the interval between 55 and 57. A zero does not imply the absence of an attribute. For instance, 00 does not imply there is no temperature at all.
Ratio level of measurement: It categorizes, ranks and specifies that the interval is constant. In this case, a zero means the absence of the attribute.
Importance of statistics in business decision making
Statistics is useful in business decision making. Statistics is applied in designing market research to provide marketing questions such as market share, competitors, among other things. Besides, it is applied in solving quantitative problems such as determining the optimal price and optimal quantity. An example of its usefulness is the management of inventory. Inventory management models such as the EOQ, among others, are statistical applications.
It is also applied to quality control in manufacturing and other business processes. Through statistical techniques, business can construct control charts, among other techniques, to determine whether a process is in control or out of control.
Business research questions
What's company position in the market?
This business research question seeks to determine the company’s product positioning about competitors in the market. It involves a survey to determine the firm's market share, approving and negative attributes for the product and competitors. Statistics will be useful in designing and conducting the study as well as analyzing, summarizing, interpreting and representing the data.
Why are some people not buying our products?
A business can conduct a survey to determine the factors that make some customers not to buy its products. It could be designed using open-ended questions for clients to indicate the attributes they don't admire in the product.
References
Jani, P. (2014). Business statistics (1st ed., pp. 1-8). Delhi: PHI Learning.
Brown, R. & Saunders, M. (2008). Dealing with statistics (1st ed., pp. 1-6). Maidenhead:
McGraw Hill, Open University Press.