You are on page 1of 56

# Lesson 4

Graph Capabilities

Agenda

Pareto Chart
Histogram
Normality Test
Marginal Plot
Box Plot
Scatter Plot
Matrix Plot

 The graphical tools of Minitab are used across
organizations to understand and interpret data for
taking key business decisions. Some examples that
are used in the following slides include:
• Analyze defects data to identify areas where most defects are originating
from. This is also called distinguishing “vital few" from the "trivial many” so
that you can focus your emphasis on resolving the vital few.
• Analyze the shape and spread of data to understand how large is the
variation in the data.
• Conduct Normality tests to identify whether the data distribution is Normal or
Non Normal
• Assess relationship between two or more variables to see if one can impact
another

Graphs

PARETO GRAPH

Steps to create a Pareto Chart

A Pareto Chart is used to
determine which of the defects
comprise the "vital few" and
which are the "trivial many”.

Let’s generate a Pareto Chart
from an existing set of data.
1. Go to File> Open Project
2. Load an existing file to load the
set of data. In this example we

Steps to create a Pareto Chart To generate Pareto Chart go to Stat > Quality Tools> Pareto Charts .

Steps to create a Pareto Chart Fill out the fields as follows: •Defects or Attribute data are all types of errors that are identified in the given data this example we will select “Category” for Defects. In this example we will select “Quantity”. •Click Ok to generate the chart. •Frequencies are the number of defects according to category in the given data. .

place the cursor in the field and then double click the column name as it appears in the left side of the dialog box. Tip 2 – You can add Title to the Chart and labels to the x and y axis by using the Options… .Tips Tip 1 – To select a column for a particular field.

.Steps to create a Pareto Chart Minitab generates the Pareto Chart. This can be copied into PowerPoint or saved as a graphic file for future use. F and partly in E. This makes it easier for us to focus our efforts towards resolving few problem areas rather than focusing on a large set of categories. Note that 80% of the defects lie in the category of G.

.Save the Graph for future use Go to File > Save Graph As… to save the graph as a file for future use.

Graphs HISTOGRAM .

.mpj to analyze the Grades of students. •Go to File > Open Project… •Load the file.Steps to create a Histogram Histogram is used to examine the shape and spread of sample data. •Now let’s generate the Histogram of the GPA results. Let’s generate a Histogram from an existing set of data. In this example we use GPA.

generate the Histogram by Graph > Histogram .Steps to create a Histogram Once the data is loaded.

” This type of Histogram is used to evaluate how well the sample data follows a specific distribution. •Click on “With Fit” to select the type of histogram you want to create. .Steps to create a Histogram The most commonly used Histogram type is “With Fit. •Then click OK.

Double click on GPA to select.Steps to create a Histogram Fill out the next screen as follows: •We will select GPA as our X value for our Graph Variables. • Click OK .

. We can use Minitab to test this distribution for normality.Steps to create a Histogram Minitab now completes our histogram for us ready to be copied and pasted into your PowerPoint presentation or save for later use. This data does not look like it is very normal.

Histogram NORMALITY TEST .Graphs .

•Let’s test the normality of the GPA data as used for creating Histograms. •Data is considered Normal if the P value of the data is >= 0.05.Steps to conduct Normality Test in Histogram •Normality Test is conducted to test if the data in the distribution is Normal. To start go to: •Stat > Basic Statistics > Graphical Summary .

Steps to conduct Normality Test in Histogram Fill out the screen as follows: •Double Click GPA to select it as the variable for Normality Test. •Click Ok Tip: Use “By variables” to create separate graphical summaries for each level of a grouping variable. The “Confidence Level” indicates the level of confidence to which the normality of data will be calculated. .

05. So.038 which is less than 0. . Also note that P value of our data is .Steps to conduct Normality Test in Histogram Note that now we not only have our Histogram but a number of other descriptive statistics as well. we conclude with 95% confidence that the data is not normal.

Graphs .Histogram MARGINAL PLOT .

Boxplots . Load the MPJ file by File > Open project . The below file includes data for tests conducted on 6 types of bottle heads across two shifts to see how much quantity of Filler was being entered. or Dotplots in the margins of the x.and y-axes. .Steps to create Marginal Plots Marginal plots are used to assess the relationship between two variables and examine their distributions. A marginal plot is like a scatter plot built along with Histograms .

Steps to create Marginal Plots To generate a Marginal Plot go to: •Graph > Marginal Plot… .

•Click on Ok . Please select the one you want to create.Steps to create Marginal Plots You can create Marginal Plots using Histograms. Box Plots or Dot plots. •In this example we are creating a Marginal Plot with Histogram.

•Select head for the X Variable •Select OK. This will help us analyze the relationship between kind of head and the amount of Filler.Steps to create Marginal Plots Fill out the screen as follows: •Select filler 1 for the Y variable. Tip: You can select more options on the graph and its axes by using the buttons below. .

Note that head number 1 seems to be the source of the highest reading.Steps to create Marginal Plots Note that now we not only have our Histogram but a dot plot of each head data as well. This means that Head 1 is allowing more Filler to be entered into the bottles. .

Graphs BOX PLOTS .

Steps to create Box Plots Box plot is used to assess and compare sample distributions. Box plots are useful to identify the range of the distribution and the outliers in the data. Let’s use the same data to create a Box plot. .

Steps to create Box Plots You can create different types of Box plots – with or without groups or with multiple Y axes. Select the graph and click Ok. . In this example we will create the Simple Box Plot graph.

We have taken Filler 1 as the field for generating Box Plots. Tip: You can customize your Box Plot by using the various buttons available in this dialog box. .Steps to create Box Plots Select the field on which you need to generate the Box plot.

The center line in the box is the Median of the data. Half the points are above and half of them are below the median. Any outliers are represented by * in the graph .Steps to create Box Plots Upper Whisker Median Lower Whisker The Box Plot is generated. The upper whisker indicates the upper most point in the data and the lower whisker indicates the lowest point.

To create group Box Plot. . select the With Groups Box Plot option and Click OK.Steps to create Group Box Plots Group Box Plots represent more than one type of data in the Box Plots.

. •Select “head” for our X Variable. •Click OK. Fill out the screen as follows: •Select “filler 1” for our Y variable.Steps to create Group Box Plots .

Steps to create Group Box Plots Note that now we have a box plot broken out by each of the various heads. . Note that head number 1 again seems to be the source of the high readings.

Graphs SCATTER PLOTS .

•Go to File>Open Project.mpj. •Load the GPA.Steps to create Scatter Plots A scatter plot is used to illustrate the relationship between two variables by plotting one against the other We will use the Grade Point average scores of students as an example to generate Scatter Plot. .

.Steps to create Scatter Plots Go to Graph > Scatter Plot...

connects and groups. •Let us start with creating a Simple scatter plot.Steps to create Scatter Plots •Scatter Plots can be made with multiple combinations of lines. . •Select Simple and then click OK.

Steps to create Scatter Plots Fill out the screen as follows: •Select GPA for Y Variable. . •Select Math and Verbal for our X Variables. •Click Ok.

We can display this better by tiling the graphs.Steps to create Scatter Plots We now have two Scatter plots of the data stacked on top of each other. .

Steps to create Scatter Plots To see multiple graphs as Tiles go to Window > Tile .

the overall GPA also increases.Steps to create Scatter Plots Now we can see both Scatter plots of the data as Tiles. . As the scores in each of these increases. You can see that there seems to be a positive relationship between score in Maths and GPA and the scores in Verbal and GPA.

Graphs MATRIX PLOTS .

To create Matrix Plots go to Graph > Matrix Plot. We will create Matrix plot using the same set of GPA data.Steps to create Matrix Plot Matrix Plot helps to assess the relationships between many pairs of variables at once by creating an array of scatter plots. .

Select the type of graph and click OK.Steps to create Matrix Plot Let us create a Matrix Plot with Groups. .

•Click OK.Steps to create Matrix Plot Fill out the fields as follows: •Click in the “Graph variables” block •Highlight all three available data sets •Click on the “Select” button. .

.Steps to create Matrix Plot We now have a series of Scatter plots. Note that there appears to be a strong correlation between Verbal and GPA data. each one corresponding to a combination of the data sets available.

 Conduct Normality tests to identify whether the data distribution is Normal or Non Normal  Assess relationship between two or more variables to see if one can impact another We learnt about important graphs like  Pareto Chart  Histogram  Normality Test  Marginal Plot  Box Plot  Scatter Plot. and  Matrix Plot We also learnt how these graphs are created in Minitab and how these can help in business scenarios.Summary In this session we learnt about common business scenarios where Minitab is extensively used. . These include:  Analyze defects data to identify areas where most defects are originating from. In the next session we will cover statistical capabilities of Minitab.  Analyze the shape and spread of data to understand how large the variation in the data is. This is also called distinguishing “vital few" from the "trivial many” so that you can focus your emphasis on resolving the vital few.

Test Questions 1. In the Pareto chart defects or attribute data field defines a) Types or categories of defects in the data b) no of times data is present c) both a & b d) none of the above Answer: a) Types or categories of defects in the data .

In the normality test data is considered not normal if the p value is a) equal to 0.Test Questions 2.05 d) none Answer: b) less than 0.05 c) greater than 0.05 b) less than 0.05 .

Matrix plot is used to a) illustrate the relationship between two variables b) assess the relationships between many pairs of variables c) assess the relationship between two variables and examine their distributions d) all the above Answer: b) assess the relationship between many pairs of variables .Test Questions 3.

Test Questions 4. Which menu convention we are using for hypothesis tests in Minitab? a) File menu b) Stat menu c) Tool menu d) Edit menu e) None Answer: b) Stat menu .

tiles or minimized form? a) window menu b) tool menu c) help menu d) assistant menu e) all the above Answer: a) window menu . Which menu convention is used to arrange the windows as a cascade.Test Questions 5.

connects and groups . d) None Answer: c) Multiple combinations of lines. Scatter Plots can be made with a) Multiple groups b) Multiple lines c) Multiple combinations of lines. connects and groups.Test Questions 6.

Test Questions 7. In the Box Plot the center line in the box defines a) median of the data b) upper most point c) lowest Point d) none Answer: a) Median of the data .

End Of Session .4 .