Читать книгу Statistics and Probability with Applications for Engineers and Scientists Using MINITAB, R and JMP - Bhisham C. Gupta, Irwin Guttman - Страница 55

Bar charts are commonly used to describe qualitative data classified into various categories based on sector, region, different time periods, or other such factors. Different sectors, different regions, or different time periods are then labeled as specific categories. A bar chart is constructed by creating categories that are represented by labeling each category and which are represented by intervals of equal length on a horizontal axis. The count or frequency within the corresponding category is represented by a bar of height proportional to the frequency. We illustrate the construction of a bar chart in the examples that follow.

Example 2.4.3 (Companies' revenue) The following data give the annual revenues (in millions of dollars) of five companies A, B, C, D, and E for the year 2011:

78, 92, 95, 94, 102

Construct a bar chart for these data.

Solution: Following the previous discussion, we construct the bar chart as shown in Figure 2.4.3.

Figure 2.4.3 Bar chart for annual revenues of five companies for the year 2011.

Example 2.4.4 (Auto part defect types) A company that manufactures auto parts is interested in studying the types of defects in parts produced at a particular plant. The following data shows the types of defects that occurred over a certain period:

2	1	3	1	2	1	5	4	3	1	2	3	4	3	1	5	2	3	1	2	3	5	4	3	1
5	1	4	2	3	2	1	2	5	4	2	4	2	5	1	2	1	2	1	5	2	1	3	1	4

Construct a bar chart for the types of defects found in the auto parts.

Solution: In order to construct a bar chart for the data in this example, we first need to prepare a frequency distribution table. The data in this example are the defect types, namely 1, 2, 3, 4, and 5. The frequency distribution table is shown in Table 2.4.2. Note that the frequency distribution table also includes a column of cumulative frequency.

Now, to construct the bar chart, we label the intervals of equal length on the horizontal line with the category types of defects and then indicate the frequency of observations associated with each defect by a bar of height proportional to the corresponding frequency. Thus, the desired bar graph, as given in Figure 2.4.4, shows that the defects of type 1 occur the most frequently, type 2 occur the second most frequently, and so on.

Table 2.4.2 Frequency distribution table for the data in Example 2.4.4.

		Frequency	Relative	Cumulative
Categories	Tally	or count	frequency	frequency
1	///// ///// ////	14	14/50	14
2	///// ///// ///	13	13/50	27
3	///// ////	9	9/50	36
4	///// //	7	7/50	43
5	///// //	7	7/50	50
Total		50	1.00

Figure 2.4.4 Bar graph for the data in Example 2.4.4.

MINITAB

Using MINITAB, the bar chart is constructed by taking the following steps.

1 Enter the category in column C1.

2 Enter frequencies of the categories in C2.

3 From the Menu bar select Graph Bar Chart. This prompts the following dialog box to appear on the screen:

4 Select one of the three options under Bars represent, that is, Counts of unique values, A function of variables, or Values from a table, depending upon whether the data are sample values, functions of sample values such as means of various samples, or categories and their frequencies.

5 Select one of the three possible bar charts that suits your problem. If we are dealing with only one sample from a single population, then select Simple and click OK. This prompts another dialog box, as shown below, to appear on the screen:

6 Enter C2 in the box under Graph Variables.

7 Enter C1 in the box under Categorical values.

8 There are several other options such as Chart Option, scale; click them and use them as needed. Otherwise click OK. The bar chart will appear identical to the one shown in Figure 2.4.4.

USING R

We can use built in ‘barplot()’ function in R to generate bar charts. First, we obtain the frequency table via the ‘table()’ function. The resulting tabulated categories and their frequencies are then inputted into the ‘barplot()’ function as shown in the following R code.

DefectTypes = c(2,1,3,1,2,1,5,4,3,1,2,3,4,3,1,5,2,3,1,2,3,5,4,3, 1,5,1,4,2,3,2,1,2,5,4,2,4,2,5,1,2,1,2,1,5,2,1,3,1,4) #To obtain the frequencies counts = table(DefectTypes) #To obtain the bar chart barplot(counts, xlab=‘Defect type’, ylab=‘Frequency’)