Response table for Analyze Taguchi
Design
Learn more about Minitab 18
Find definitions and interpretation guidance for every statistic in the Response table.
Use the response tables to select the best level for each factor. Usually you have the
following objectives with a Taguchi design:
Minimize the standard deviation
Maximize the S/N ratio
Meet a target with the mean (static design)
Meet a target with the slope (dynamic design)
Use the delta and rank values to identify the factors that have the largest effect on each
response characteristic. Then, determine which levels of these factors meet your
objectives. Sometimes, the best level of a factor for one response characteristic is
different from the best level for a different response characteristic. To resolve this issue,
it may help to predict the results for several combinations of factors levels to see which
one produces the best result.
Factorial plots and scatterplots
for Analyze Taguchi Design
Learn more about Minitab 18
Find definitions and interpretation guidance for every plot.
In This Topic
Main effects plot
Interaction plot
Scatterplot
Main effects plot
Main effects plots show how each factor affects the response characteristic (S/N ratio,
means, slopes, standard deviations). A main effect exists when different levels of a factor
affect the characteristic differently. For a factor with two levels, you may discover that
one level increases the mean compared to the other level. This difference is a main
effect.
In these results, the main effects plot for S/N ratio indicates that Fertilizer has the largest effect on the
signal-to-noise ratio. On average, experimental runs with Fertilizer 2 had much higher signal-to-noise
ratios than experimental runs with Fertilizer 1. Spraying had a small effect or no effect on the signal-
to-noise ratio.
Interaction plot
Taguchi designs traditionally focus on main effects, but it is important to test suspected
interactions. Use the interaction plots to determine whether the effect of one factor on a
response characteristic (S/N ratio, means, slopes, or standard deviations) depends on the
level of another factor.
Interpretation
The interaction plots show interactions between factors. Minitab creates an interaction
plot by plotting for two factors the characteristic average for each combination of factor
levels. So, for two factors with two levels each, Minitab plots four points that represent
the four possible combinations. The levels of one factor are indicated on the horizontal
axis, whereas the levels of the other factor are indicated by different colored lines and
symbols.
If the lines are parallel to each other, then there is no interaction between the two
factors.
If the lines are not parallel to each other, then there is an interaction between the
two factors.
In these results, for the S/N ratios, the lines are close to parallel. Variety 2 has a higher S/N ratio than
Variety 1 using both Fertilizer 1 and 2.
In addition to the interaction plots, examine the linear model analysis to determine
whether the interaction is significant.
Scatterplot
The scatterplots for dynamic response experiments show the responses plotted against
the signal. Each plot shows all the data for one setting of the control factors in the
experiment. The following items are shown on the plot:
The least squares regression line through the reference point.
The row number at the top of each plot, which refers to the first row in which the
factor settings for that plot appear.
The signal-to-noise ratio, slope, and standard deviation for the factor settings,
which are located at the bottom of the plot.
Interpretation
The graphs are arranged in decreasing order of the signal-to-noise ratio, so that the
experimental runs with the highest ratios are plotted first. If the experiment has more
than nine combinations of factor settings, Minitab displays more than one graph window
of scatterplots.
Consider the following questions:
Does the fitted line conform closely to the pattern of the data for the "best"
experimental runs?
Is the pattern of the data a straight line or is it curved?
If it is a line, does it follow the fitted line, or is it shifted?
Are there any unusual response values or outliers?
In this plot, there is a substantial difference in the spread of the data between the best and worst fits.
For example, in the plot in the first cell, for row 21, the data are very close to the line. In the plot in the
lower left corner, for row 9, the data vary much more widely. The standard deviation for row 21 is
0.4089, but is much larger in row 9. The standard deviation in row 9 is 1.1718.
Interpretation
Average response characteristics
For each factor, Minitab calculates the average of the response characteristic at
each level of the factor. For example, the design includes factor A at 2 levels (1
and 2) and 4 measurements at each level. Minitab calculates the mean of the 4
S/N ratios at level 1 and the mean of the other 4 S/N ratios at level 2. For more
information, go to Methods and formulas for Analyze Taguchi Design.
Signal-to-Noise Ratio
Minitab calculates a separate signal-to-noise ratio (S/N) for each combination of
control factor levels in the design. You can choose from different S/N ratios,
depending on the goal of your experiment. In all cases, you want to maximize the
S/N ratio.
Means (for static designs)
Minitab calculates a separate mean for each combination of control factor levels
in the design.
Slopes (for dynamic designs)
Minitab calculates a separate slope for each combination of control factor levels
in the design.
Standard deviations
Minitab calculates a separate standard deviation for each combination of control
factor levels in the design.
Delta
Measures the size of the effect by taking the difference between the highest and
lowest characteristic average for a factor.
Rank
The ranks in a response table help you quickly identify which factors have the
largest effect. The factor with the largest delta value is given rank 1, the factor
with the second largest delta is given rank 2, and so on.
In these results, the response tables show the following:
For the Signal to Noise Ratios, B is ranked 1, followed by D, A, and C.
For the Means, B is ranked 1, followed by A, C, and D.
For the Standard Deviations, C is ranked 1, followed by B, A, and D.
Suppose you want to choose factors levels that minimize the standard
deviation and maximize the S/N ratio and the mean. For example, for factor B,
the average S/N ratio for all runs with level 1 is 42.15 and the average for runs
with level 2 is 34.21. This indicates that level 1 maximizes the signal-to-noise
ratio.
Response Table for Signal to Noise Ratios
Larger is better
Level Material Diameter Dimples Thickness
1 41.62 42.15 41.16 34.70
2 34.75 34.21 35.20 41.66
Delta 6.87 7.93 5.96 6.96
Rank 3 1 4 2
Response Table for Means
Level Material Diameter Dimples Thickness
1 147.26 161.70 133.65 87.56
2 73.54 59.10 87.15 133.24
Delta 73.73 102.60 46.50 45.68
Rank 2 1 3 4
Response Table for Standard Deviations
Level Material Diameter Dimples Thickness
1 6.417 7.000 7.000 6.417
2 5.250 4.667 4.667 5.250
Delta 1.167 2.333 2.333 1.167
Rank 3 2 1 4
S
S represents the standard deviation of the distance between the data values and the
fitted values. S is measured in the units of the response.
Interpretation
Use S to assess how well the model describes the response. S is measured in the units of
the response variable and represents how far the data values fall from the fitted values.
The lower the value of S, the better the model describes the response. However, a low S
value by itself does not indicate that the model meets the model assumptions. You
should check the residual plots to verify the assumptions.
For example, you work for a potato chip company that examines the factors that affect
the percentage of crumbled potato chips per container. You reduce the model to the
significant predictors, and S is calculated as 1.79. This result indicates that the standard
deviation of the data points around the fitted values is 1.79. If you are comparing
models, values that are lower than 1.79 indicate a better fit, and higher values indicate a
worse fit.
R-sq
R2 is the percentage of variation in the response that is explained by the model. It is
calculated as 1 minus the ratio of the error sum of squares (which is the variation that is
not explained by model) to the total sum of squares (which is the total variation in the
data).
Interpretation
The following fitted line plots illustrate different R 2 values. The first plot illustrates a
simple regression model that explains 85.5% of the variation in the response. The second
plot illustrates a model that explains 22.6% of the variation in the response. The more
variation that is explained by the model, the closer the data points fall to the fitted
regression line. Theoretically, if a model could explain 100% of the variation, the fitted
values would always equal the observed values and all the data points would fall on the
fitted line. However, even if R2 is 100%, the model does not necessarily predict new
observations well.
Consider the following issues when interpreting the R 2 value:
R2 always increases when you add more predictors to a model. For example, the
best five-predictor model will always have an R 2 that is at least as high as the best four-
predictor model. Therefore, R2 is most useful when you compare models of the same
size.
Small samples do not provide a precise estimate of the strength of the
relationship between the response and predictors. If you need R 2 to be more precise, you
should use a larger sample (typically, 40 or more).
R2 is just one measure of how well the model fits the data. Even when a model has
a high R2, you should examine the residual plots to verify that the model meets the
model assumptions.
R-sq (adj)
Adjusted R2 is the percentage of the variation in the response that is explained by the
model, adjusted for the number of predictors in the model relative to the number of
observations. Adjusted R2 is calculated as 1 minus the ratio of the mean square error
(MSE) to the mean square total (MS Total).
Interpretation
Use adjusted R2 when you want to compare models that have different numbers of
predictors. R2always increases when you add a predictor to the model, even when there
is no real improvement to the model. The adjusted R2 value incorporates the number of
predictors in the model to help you choose the correct model.
For example, you work for a potato chip company that examines the factors that affect
the percentage of crumbled potato chips per container. You receive the following results
as you add the predictors in a forward stepwise approach.
Mode
% Potato Cooling rate Cooking temp R2 Adjusted R2
l
1 X 52% 51%
2 X X 63% 62%
3 X X X 65% 62%
The first model yields an R2 of more than 50%. The second model adds cooling rate to
the model. Adjusted R2 increases, which indicates that cooling rate improves the model.
The third model, which adds cooking temperature, increases the R 2 but not the adjusted
R2. These results indicate that cooking temperature does not improve the model. Based
on these results, you consider removing cooking temperature from the model.
Coefficients table for Analyze Taguchi
Design
Learn more about Minitab 18
Find definitions and interpretation guidance for every statistic in the Estimated Model
Coefficients table.
In This Topic
Coef
SE Coef
T-value
P-Value
Coef
The coefficient describes the size and direction of the relationship between a term in the
model and the response variable. The absolute value of the coefficient indicates the
relative strength of each factor. To minimize multicollinearity among the terms, the
coefficients are all in coded units.
The number of coefficients Minitab calculates for a factor is the number of levels minus
one. If a factor has 3 levels, Minitab provides 2 coefficients, which correspond to factor
levels 1 and 2. If a factor has 2 levels, Minitab provides 1 coefficient, which corresponds
to factor level 1. Minitab includes the values or text that correspond to the level.
Interpretation
In Taguchi designs, the magnitude of the factor coefficient usually mirrors the factor rank
in the response table. Depending on your analysis, the response can be a signal-to-noise
ratio, the mean for a static design, the slope for a dynamic design, or a standard
deviation.
The size of the effect is usually a good way to assess the practical significance of the
effect that a term has on the response variable. The size of the effect does not indicate
whether a term is statistically significant because the calculations for significance also
consider the variation in the response data. To determine statistical significance, examine
the p-value for the term.
SE Coef
The standard error of the coefficient estimates the variability between coefficient
estimates that you would obtain if you repeated the same experiment again and again.
The calculation assumes that the experimental design and the coefficients to estimate
would remain the same if you sampled again and again.
Interpretation
Use the standard error of the coefficient to measure the precision of the estimate of the
coefficient. The smaller the standard error, the more precise the estimate. Dividing the
coefficient by its standard error calculates a t-value. If the p-value associated with this t-
statistic is less than your significance level, you conclude that the coefficient is
statistically significant.
T-value
The t-value measures the ratio between the coefficient and its standard error.
Interpretation
Minitab uses the t-value to calculate the p-value, which you use to test whether the
coefficient is significantly different from 0.
You can use the t-value to determine whether to reject the null hypothesis. However, the
p-value is used more often because the threshold for the rejection of the null hypothesis
does not depend on the degrees of freedom. For more information on using the t-value,
go to Using the t-value to determine whether to reject the null hypothesis.
P-Value
The p-value is a probability that measures the evidence against the null hypothesis.
Lower probabilities provide stronger evidence against the null hypothesis.
Interpretation
To determine whether a coefficient is statistically different from 0, compare the p-value
for the term to your significance level to assess the null hypothesis. The null hypothesis is
that the coefficient equals 0, which implies that there is no association between the term
and the response characteristic that you select. In Taguchi designs, response
characteristics refer to functions of the response, such as means, standard deviations,
slopes, and signal-to-noise ratios.
Usually, a significance level (denoted as α or alpha) of 0.05 works well. A significance
level of 0.05 indicates a 5% risk of concluding that the coefficient is not 0 when it is.
Frequently, a significance level of 0.10 is used for evaluating terms in a model.
P-value ≤ α: The association is statistically significant
If the p-value is less than or equal to the significance level, you can conclude that
there is a statistically significant association between the response characteristic
and the term.
P-value > α: The association is not statistically significant
If the p-value is greater than the significance level, you cannot conclude that
there is a statistically significant association between the response characteristic
and the term. You may want to refit the model without the term.
If there are multiple predictors without a statistically significant association with
the response, you can reduce the model by removing terms one at a time. For
more information on removing terms from the model, go to Model reduction
