Unit 5: How a Randomized Block Design Differs From a One-Way ANOVA

Size: px
Start display at page:

Download "Unit 5: How a Randomized Block Design Differs From a One-Way ANOVA"

Transcription

1 Minitab Notes for STT 6305: nalysis of Variance Models Department of Statistics and iostatistics SU East ay Unit 5: How a Randomized lock Design Differs From a One-Way NOV 5.1. First Dataset Real Estate ppraisal In this unit we look at two datasets that illustrate randomized block designs. In the first dataset, three real estate appraisers independently examined each of five properties chosen at random from a particular neighborhood and gave appraised values. The manager of the appraisal service that employs them suspects that systematic differences among their methods of appraisal are causing undesirable inconsistencies in appraised values provided by her service. The issue is whether there are, in fact, systematic differences among the values reported by the three appraisers. The data, appraised values in thousands of dollars, are shown below. PPRISED VLUES (In thousands of dollars) ppraiser Property The data are from anavos and Miller: n Introduction to Modern usiness Statistics, Duxbury, 1993, page 450. The design of this study is a straightforward generalization of the paired-data design we studied in Unit 4. The change is that instead of making a comparison between the bids by two auto repair garages we are making comparisons among the appraisals of three real estate appraisers. In Unit 4 each car produced a pair of bids. Here each property yields a triplet of appraised values. In the language of experimental design, the generalization of pair is block. Here the properties play the role of blocks. Problems ut these data (between dotted lines) from above and paste them directly into columns c1 ('Prop'), c2 (''), c3 (''), and c4 ('') of a Minitab worksheet. (When you paste, put the cursor in row 1 of c1 and use spaces as delimiters.) The property number should go into c1 and the values from appraisers,, and into columns c2, c3, and c4, respectively. Label these columns appropriately. Next, stack the data "in blocks of columns" as shown below. ompare the contents of columns c1 through c4 with the contents of columns c11 through c13. Make a printout of columns c11 through c13. DT > Stack > locks, use column names as subscripts. MT > stack (c2 c1) (c3 c1) (c4 c1) (c11 c12); SU> subs c13; SU> usenames.

2 No menu item; type column names into worksheet MT > name c11 'Value' c12 'Property' c13 'pprsr' DT Display data MT > print c11 c12 c13 Row Value Property pprsr Minitab Notes for STT 6305 lock Design & One Factor NOV Unit 5-2 Notice that c13 has "text" data rather than "numeric" data. Many Minitab procedures can use text data to designate groups, but some of the procedures that originated in older releases require numerical data. (You can use the patterned data procedure to put appraiser numbers 1, 2, 3 into c14. lternatively, you can use the menu path DT > ode > Text to numeric to convert the letters in c13 into numbers in c14. all c14 'pprn'.) Using R. Use the statements below to make the data available in R and print them as a dataframe. Notice that we designate property and appraiser as factor variables from the start. value = c(90, 94, 91, 85, 88, 93, 96, 92, 88, 90, 92, 88, 84, 83, 87) prop = as.factor(rep(1:5, times=3)) apprsr = as.factor(rep(c("", "", ""), each=5)) ppr.val = data.frame(value, prop, apprsr); ppr.val 5.2. Exploratory Methods The sample means of appraised values differ somewhat among the appraisers: 89.6 for, 91.8 for, 86.8 for. (See Problem ) This pattern of the appraised values may show that the differences in sample means reflect actual differences in methodology or standards among the appraisers (i.e., differences in population means). On the other hand, the differences in sample means may be due to chance. The question is which real differences or chance alone? ecause these means do not take into account that each property was evaluated by all three appraisers, the means by themselves do not provide an adequate way to answer the question. For the same reason that two dotplots on the same scale were inappropriate in Unit 4, three dotplots on the same scale would be a poor way to graph the data here. Here we cannot resort to looking at one column of differences as we did in Unit 4. However, an appropriately labeled scatterplot of the data can provide some useful insights (next page). GRPH > Scatterplot > With Groups, Labels, Data labels from columns This plot displays the blocked nature of the data. It is clear that ppraiser gave the highest value for all 5 properties, while ppraiser gave the lowest value for all but one. meaningful pattern of differences does seem to emerge.

3 Minitab Notes for STT 6305 lock Design & One Factor NOV Unit 5-3 Scatterplot of Value vs Property pprsr 92 Value Property 4 5 Problems Find the means for each appraiser with the command table c13; subcommand means c11. or with the command describe c11; subcommand by c13.) Use the command table 'pprsr' 'Property'; with subcommand means 'Value'. to make a table of the data with marginal means Using R. Use the R code below (one block at a time) to make three graphical displays of the data. The first two are also exercises in data types, so explain their use. The interaction plot is intended for use with slightly more advanced designs but serves well here with appropriate labeling; it is programmed to handle factor variables directly. colr = c("red", "blue", "darkgreen") plot(as.numeric(prop), value, pch=as.character(apprsr), col=colr[as.numeric(apprsr)], xlab="property", ylab="value", main="ppraised Values by Property and ppraiser") plot(0, xlim=c(1,5), ylim=c(80,100), type="n", xlab="property", ylab="value", main="ppraised Values by Property and ppraiser") # set up but no plot char = unique(apprsr) for (i in 1:3) { lines(value[as.numeric(apprsr)==i], type="b", col=colr[i], pch=as.character(unique(apprsr))[i]) } # no x-variable given, indices 1:5 used as x interaction.plot(prop, apprsr, value, xlab="property", ylab="value", main="ppraised Values by Property and ppraiser") Using R. Display three boxplots using plot(apprsr, value) and also using boxplot(value ~ apprsr). What do you get with plot(as.numeric(apprsr), value)?

4 5.3. nalysis of Variance Minitab Notes for STT 6305 lock Design & One Factor NOV Unit 5-4 The NOV procedure for analyzing this block design is shown below. (Either numeric or text subscripts for appraisers would work.) MT > anova c11 = c13 c12; SU> rand c12; SU> rest; SU> resid c15; SU> ems. NOV: Value versus pprsr, Property Factor Type Levels Values pprsr fixed 3,, Property random 5 1, 2, 3, 4, 5 nalysis of Variance for Value Source DF SS MS F P pprsr Property Error Total S = R-Sq = 83.71% R-Sq(adj) = 71.49% Expected Mean Square Variance Error for Each Term using Source component term restricted model) 1 pprsr 3 (3) + 5 Q[1] EMS Notation: Q[] = θ, () = σ 2, and 2 Property (3) + 3 (2) number inside [] or () shows Source 3 Error (3) It is clear from this printout that there are significant differences among ppraisers. The F-ratio is F = MS(pp)/MS(Err) = /3.983 = The P-value is the area under the density curve of the distribution F(2, 8) that lies to the right of 7.88, which is Thus, the ppraiser effect is significant at the 5% level, but not quite significant at the 1% level. It is also worthwhile to test the Property effect. y happenstance, the P-value of this test is about the same as for the ppraiser effect. Of course, it is no surprise that the properties are of different values. It is precisely the suspicion that they might be highly variable that lead us to treat Properties as blocks in the design. learly, this was a useful choice of design. The table of expected mean squares is based on the model Y ij = µ + α i + j + e ij, where i = 1, 2, 3; j = 1,..., 5; Σ i α i = 0; j iid N(0, σ 2 ); and e ij iid N(0, σ 2 ). In Minitab's Expected Mean Square (EMS) tables, the terms indicated by parentheses ( ) are variances, where the number inside the parentheses gives the appropriate row label. The terms Q[ ] are quadratic forms involving constants. Specifically for our design, EMS(pp) = σ 2 + 5[Σ i α i 2 ]/2 = σ 2 + 5θ, EMS(Prp) = σ 2 + 3σ 2, EMS(Err) = σ 2.

5 Minitab Notes for STT 6305 lock Design & One Factor NOV Unit 5-5 Formally, the test for the ppraiser effect is the test of the null hypothesis H 0 : θ = 0 against the alternative H a : θ > 0. Notice that, if the null hypothesis is true, then EMS(pp) = EMS(Err) = σ 2. In these circumstances (and, of course, subject to the conditions of normality and equal cell variances specified in the model), the ratio MS(pp)/MS(Err) has an F-distribution with 2 and 8 degrees of freedom. Roughly speaking, because the two MSs then estimate the same quantity σ 2, their ratio ought to be about 1. Thus values very much larger than 1 lead us to suspect that θ is not equal to 0. Similarly, the test for the Property effect is the test of the null hypothesis H 0 : σ 2 = 0 against the alternative H a : σ 2 > 0. If this null hypothesis is true, then EMS(Prp) = EMS(Err), and so it is natural to form the ratio MS(Prp)/MS(Err) to test this null hypothesis in much the same way as we tested for the ppraiser effect. In this relatively simple design it turns out that the F-ratios for both tests of hypothesis have MS(Err) in the denominator. In more complex designs, it is not always appropriate to "test against the error term" in this way. In later units we shall see that the EMS table is an indispensable guide to the formation of appropriate ratios for testing hypotheses in an NOV table. Notes on using the Minitab anova procedure to analyze a block design: The subcommand restrict has minimal effect on output for a block design. If restrict is used, then in the EMS table (generated by the ems subcommand) has EMS() = EMS(prrsr) = σ 2 + bθ (denoted as (3) + 5 Q[1] in the printout above), where b is the number of blocks. ut without the subcommand restrict the printout shows EMS() = σ 2 + θ. This is not a contradictory result; it arises from two different definitions of θ. In some more complex NOV designs where one factor is fixed and another is random, it makes an important difference whether the subcommand restrict is used. We will revisit this issue later in the course. (Some texts recommend using restricted models and others do not. For the current simple design, the distinction makes no difference in the F-ratios or the conclusions drawn from them.) In practice, blocking factors are almost always random. (Some authors will not use the terminology blocking factor except to refer to a random factor.) Often a random blocking factor is introduced into the design in hopes it will "explain" enough variability that MS(Error) of the NOV table tends to become small enough to improve the power of the test for the main factor. In many texts block designs are first introduced without making a distinction whether the blocking factor is considered as a fixed or a random factor. For a simple block design as in this unit, the NOV table (including SS, MS, F, and P columns) is exactly the same whether or not the random subcommand is used. However, the Minitab EMS table shows a variance component for every random factor. So if lock (here Property) is declared as random, then we see an estimate of σ 2 in the EMS table (7.083 above). ut if lock is not declared as random, then the EMS table shows no such estimate. The R procedure in problem below makes essentially the same NOV table as the one shown above from Minitab, and it makes no EMS table. This R procedure is the same as the above NOV procedure where the blocking factor is not declared to be random. ecause the R procedure shows no EMS table, this makes no difference in the output shown. (More advanced R procedures can handle random factors.) Problems Make a normal probability plot of the residuals from this NOV model. Does it show that the data are reasonably close to normal? Perform an appropriate test of normality and give the P-value.

6 Minitab Notes for STT 6305 lock Design & One Factor NOV Unit ompare the test for the ppraiser effect with the test for the Property effect. How is it possible that the two P-values are nearly equal when the F-statistics have quite different values (7.88 vs. 6.33)? Suppose that the Property effect had not been found to be significant. Without using special statistical terminology, explain what this would mean about the values of the five properties selected onsider the variance components in Minitab's EMS table. The variance component for Error is the estimated value of σ 2, which we have already seen to be MS(Err) = The variance component for Property is the estimated value of σ 2. ut MS(Prp) estimates σ 2 + 3σ 2, not σ 2. How can the values of MS(Prp) and MS(Err) be used together to obtain a numerical estimate of σ 2? Show that your answer agrees with the variance component for Property shown in the EMS table Suppose you own 5 similar properties, consider selling them, and wonder if they should all be put on the market at the same price. Then you might consult three appraisers (randomly chosen from among the many available) for advice. In this case, how would the NOV model be changed? One advantage of an appropriately used block design with b blocks (e.g., Properties) is that it makes better use of data in the effort to detect significant differences among t treatment groups (e.g., ppraisers) than would a one-way NOV. The relative efficiency of a block design compared to a one-way NOV is defined as the ratio RE = MS(Err) OneWay /MS(Err) lock. The block design uses bt observations. The idea is that a one-way NOV would require about bt(re) total observations to do the "same job." (See, for example, page 973 in O/L 6e or hapter 14 of Snedecor/ochran for some details.) In practice, data from a legitimate one-way NOV will seldom be available. So we must estimate MS(Err) OneWay from the NOV table of the block experiment as MS(Err) OneWay [(b 1)MS(lock) + b(t 1)MS(Error)] / (bt 1). (aution: bogus MS(Err) OneWay computed as in section 4.5 below cannot be used to find relative efficiency!) side from the fact that MS(Err) OneWay can only be approximated, there are other reasons that RE can only be used as a rough guide. In our data b = 5 and t = 3. Find RE for these data. How many observations (instead of b = 5) on each of t = 3 ppraisers does your value of RE indicate we would need for a one-way NOV having a "precision" equivalent to our blocked experiment? In R, use the code below to make an NOV table. ompare with the Minitab NOV table. anova(lm(value ~ apprsr + prop)) 5.4. Relationship to One-Way NOV In Unit 4 we saw what can happen if paired structure is ignored in a comparison of two means. For the data in Unit 4, the incorrect independent 2-sample t test failed to find a large effect that was found using the correct paired t procedure. For the appraiser data, we mentioned above that comparing three dotplots of values from the three appraisers would not be effective because this comparison ignores the blocked structure of the experiment. The computational equivalent of this ineffective comparison of dot plots is to analyze the appraiser data using a one-way NOV model instead of the correct block-design.

7 Minitab Notes for STT 6305 lock Design & One Factor NOV Unit 5-7 In this section we look at what happens when the INORRET (one-way NOV with ppraisers as the only factor) is used instead of the block design. INORRET PROEDURE STT > NOV > One-way MT > onew c1 = c2 NLYSIS OF VRINE ON Value SOURE DF SS MS F p pprsr Misleading test ERROR TOTL INDIVIDUL 95 PT I'S FOR MEN SED ON POOLED STDEV LEVEL N MEN STDEV ( * ) ( * ) ( * ) POOLED STDEV = The relatively large P-value for testing the ppraiser effect (about 10%) together with the considerably overlapping confidence intervals for ppraiser means generated by this incorrect procedure might lead a beginner to conclude that there are no significant differences among appraisers. Garbage in (incorrect model) garbage out (incorrect conclusion). Of course, we will not invariably get the wrong conclusion by using the wrong model, it is just that we cannot depend on reaching the correct conclusion. One of the most frequently made mistakes in statistical analysis is to analyze what is really a block design using a one-way NOV. Notice the relationship between this NOV table and the correct one: SS(pp) = 62.8 is the same in both tables. However, SS(Error) 1WY = in the bogus one-way analysis gets split out into two parts in the NOV table for the block design; the same can be said for degrees of freedom: SS(Prp) LOK + SS(Err) LOK = = DF(Prp) LOK + DF(Err) LOK = = 12 ompared with the test for ppraiser in the one-way NOV, the F-ratio F(pr) in the correct block design has a much smaller denominator (3.98). onsequently the F-ratio is a much larger (7.88) and has a smaller P-value (0.013). The block design has "explained" some of the variance in Values by taking blocks into account. Thus the SS(Err) and MS(Err) become relatively small in the block design. The appeal of block designs lies precisely in this potential ability to reduce the fundamental variance estimate MS(Err). Problems Perform an INORRET one-way NOV using Property as the factor. ompare the results with the correct NOV table from the block design Suppose you were trying to analyze data from a block design using the limited capabilities of a spread sheet, in which the one-way NOV is the only NOV procedure available. How could you use output from the spreadsheet to put together the correct analysis?

8 Minitab Notes for STT 6305 lock Design & One Factor NOV Unit Second Dataset Transformation to Stabilize Variances n experiment was conducted to compare the effects of three Temperature levels (Low=20 o, Medium=35 o, and High=50 o ) on the time until failure of a particular kind of electronic component. ecause it is thought that different manufacturing batches of the component may have somewhat different reliabilities, three components are selected at random from each of five atches for test at the three different Temperatures. Results are as shown in the table at left below. Survival times are in months. Temperature atch Low Med High MT > print c2-c4 Data Display Row Months Temp atch To produce the table at right above, we put the 'Months' into c2, 'Temp' subscripts into c3, and 'atch' subscripts into c4. Then we printed these columns in the Session window. We analyzed these data using the appropriate block model, as follows: Made an appropriate scatterplot (labeled with letters or symbols for Temperatures) and formed a preliminary conclusion as to the possible significance of the Temperature effect. Performed the analysis of variance, testing for the significance of both Temperature (main) and atch (block) effects. Neither factor is significant at the 5% level. Stored the residuals from the correct two-factor model and made a normal probability plot of them and also plotted residuals against fitted values. }}} Probability Plot of Residuals Normal - 95% I Residuals Versus the Fitted Values (response is Months) Mean E-16 StDev N 15 D P-Value Percent Residual RESI Fitted Value 12 16

9 Minitab Notes for STT 6305 lock Design & One Factor NOV Unit 5-9 Even though there are not enough data for the formal test of normality to fail, it seems that the points (except perhaps for one) fit a curve better than a line. lso, even though artlett's test for equal variances (not shown) does not fail, the plot of residuals against fitted values seems to show that residuals become increasingly variable as lifetimes increase. From the plots alone there would not be enough evidence to say that the assumptions for NOV are not met, but it is known that lifetimes of this particular kind of component tend to have an exponential distribution rather than a normal distribution. If this is the case for our data, then nonnormality is not the only difficulty. The standard deviation of an exponential distribution is the same as its mean. So if means are different for different Temperatures, then variances are also different and the homoscedasticity assumption of the NOV model is violated. F-tests in an NOV are somewhat robust against nonnormality, but not so robust against heteroscedasticity. One can show that a log transformation "stabilizes" the variances of exponential. One takes the log of each lifetime and runs the appropriate NOV on the transformed data. (It does not matter whether one takes log base e or log base 10.) For data that are suspected to be exponential, it is common practice to run two NOVs, one on the original counts and one on their logs. If the conclusions are the same, one usually reports the results from the NOV on the original data (with a footnote that the transformed data gave the same result). If the interpretations are different and the transformed appear to fit the NOV assumptions better, then the results from the transformed data are usually taken as being correct. (Unfortunately, making sense of multiple comparisons for such transformed data takes a little extra thought.) Here is the NOV for a block design using the transformed data. It shows that the Temperature effect is significant at the 5% level. MT > let c11 = loge(c1) MT > name c11 'LogMo' STT > NOV alanced MT > anova c11 = c2 c3; SU> random c3; SU> restrict; SU> resids c5; SU> ems. NOV: LogMo versus Temp, atch Factor Type Levels Values Temp fixed 3 1, 2, 3 atch random 5 1, 2, 3, 4, 5 nalysis of Variance for LogMo Source DF SS MS F P Temp atch Error Total S = R-Sq = 73.65% R-Sq(adj) = 53.88% Expected Mean Square for Each Term (using Variance Error restricted Source component term model) 1 Temp 3 (3) + 5 Q[1] 2 atch (3) + 3 (2) 3 Error (3)

10 Problems Minitab Notes for STT 6305 lock Design & One Factor NOV Unit Put the component lifetime data into a Minitab worksheet. Then use Minitab's ability to do crosstabulations to obtain a table of the data suitable for presentation to human readers, similar to the first table shown above (left side). Use the command table 'Temp' 'atch'; with subcommand data 'Months'.) Make a labeled scatterplot of the original lifetime data similar to the one for the appraiser data in Section 4.2. Perform an NOV for a block design on the original lifetimes. Make a plot of residuals against fits. Interpret the results The NOV for the block design on the log-transformed data is shown in this section. Perform it for yourself, making a normal probability plot of the residuals. Does this plot seem more nearly linear than the one shown in this section for the original data? Make a plot of residuals against fits and compare it with the one in the previous problem. Finally, did transforming the data change your interpretation of the data (as to significant differences among the Temperatures at the 5% level)? What would you say about your findings in a report intended for nonstatisticians? Perform Fisher's LSD and Tukey's HSD by hand. Use the transformed data. However, notice that a difference in logarithms (as in these multiple comparisons) is the logarithm of a ratio. How can you use this fact to compare lifetimes in months (original data scale) for different Temperatures? For the Fisher procedure use the formula LSD = t*[2ms(error) / b] 1/2, where b is the number of batches (the number of observations averaged to get each of the three Temp-level means), and t* cuts off 2.5% from the upper tail of the t distribution with df(error) = (a 1)(b 1) degrees of freedom. For the Tukey procedure use HSD = q*[ms(error) / b] 1/2, where q* can be obtained from tables of the studentized range or with the R function qtukey(.95, a, (a 1)*(b - 1)). This is for 95% confidence in the pattern of differences shown in the three Is. Why might you want to look at the HSD for 90% confidence in this particular case? Verify your results for Tukey's HSD by using Minitab's glm procedure using the menu path STT > NOV > GLM or the commands below. MT > GLM 'LogMo' = Temp atch; SU> Random 'atch'; SU> Pairwise Temp; SU> Tukey. In the Minitab's version of HSD, you have the choice of output with Is or tests (or both). Each I has total length 2HSD, centered at the appropriate difference in Temp-level means. (The Minitab 14 GLM procedure does not find Fisher LSD Is. However, notice that the quantity labeled 'SE of Difference' in the output for the Tukey HSD tests is [2MS(Error) / b] 1/2, not [MS(Error) / b] 1/2.) Minitab Notes for Statistics 6305: NOV Models by ruce E. Trumbo, Department of Statistics, SU East ay, East ay, opyright 1991, 2011 by ruce E. Trumbo. ll rights reserved. Partial support for the 1991 version from NSF grant USE The current version with Minitab professional graphics and additional examples is a draft. For comments, errata, selected answers, related materials, and permission to use beyond SU East ay please bruce.trumbo@csueastbay.edu or eric.suess@csueastbay.edu.

Unit 8: A Mixed Two-Factor Design

Unit 8: A Mixed Two-Factor Design Minitab Notes for STAT 6305 Dept. of Statistics CSU East Bay Unit 8: A Mixed Two-Factor Design 8.1. The Data We use data quoted in Brownlee: Statistical Theory and Methodology in Science and Engineering,

More information

Unit 8: A Mixed Two-Factor Design

Unit 8: A Mixed Two-Factor Design Minitab Notes for STAT 3503 Dept. of Statistics CSU Hayward Unit 8: A Mixed Two-Factor Design 8.1. The Data We use data quoted in Brownlee: Statistical Theory and Methodology in Science and Engineering,

More information

1 Introduction to Minitab

1 Introduction to Minitab 1 Introduction to Minitab Minitab is a statistical analysis software package. The software is freely available to all students and is downloadable through the Technology Tab at my.calpoly.edu. When you

More information

9 One-Way Analysis of Variance

9 One-Way Analysis of Variance 9 One-Way Analysis of Variance SW Chapter 11 - all sections except 6. The one-way analysis of variance (ANOVA) is a generalization of the two sample t test to k 2 groups. Assume that the populations of

More information

BIOL Biometry LAB 6 - SINGLE FACTOR ANOVA and MULTIPLE COMPARISON PROCEDURES

BIOL Biometry LAB 6 - SINGLE FACTOR ANOVA and MULTIPLE COMPARISON PROCEDURES BIOL 458 - Biometry LAB 6 - SINGLE FACTOR ANOVA and MULTIPLE COMPARISON PROCEDURES PART 1: INTRODUCTION TO ANOVA Purpose of ANOVA Analysis of Variance (ANOVA) is an extremely useful statistical method

More information

MORE ON MULTIPLE REGRESSION

MORE ON MULTIPLE REGRESSION DEPARTMENT OF POLITICAL SCIENCE AND INTERNATIONAL RELATIONS Posc/Uapp 816 MORE ON MULTIPLE REGRESSION I. AGENDA: A. Multiple regression 1. Categorical variables with more than two categories 2. Interaction

More information

ANOVA: Analysis of Variation

ANOVA: Analysis of Variation ANOVA: Analysis of Variation The basic ANOVA situation Two variables: 1 Categorical, 1 Quantitative Main Question: Do the (means of) the quantitative variables depend on which group (given by categorical

More information

INTRODUCTION TO ANALYSIS OF VARIANCE

INTRODUCTION TO ANALYSIS OF VARIANCE CHAPTER 22 INTRODUCTION TO ANALYSIS OF VARIANCE Chapter 18 on inferences about population means illustrated two hypothesis testing situations: for one population mean and for the difference between two

More information

Examination paper for TMA4255 Applied statistics

Examination paper for TMA4255 Applied statistics Department of Mathematical Sciences Examination paper for TMA4255 Applied statistics Academic contact during examination: Anna Marie Holand Phone: 951 38 038 Examination date: 16 May 2015 Examination time

More information

SMAM 314 Computer Assignment 5 due Nov 8,2012 Data Set 1. For each of the following data sets use Minitab to 1. Make a scatterplot.

SMAM 314 Computer Assignment 5 due Nov 8,2012 Data Set 1. For each of the following data sets use Minitab to 1. Make a scatterplot. SMAM 314 Computer Assignment 5 due Nov 8,2012 Data Set 1. For each of the following data sets use Minitab to 1. Make a scatterplot. 2. Fit the linear regression line. Regression Analysis: y versus x y

More information

Unit 12: Analysis of Single Factor Experiments

Unit 12: Analysis of Single Factor Experiments Unit 12: Analysis of Single Factor Experiments Statistics 571: Statistical Methods Ramón V. León 7/16/2004 Unit 12 - Stat 571 - Ramón V. León 1 Introduction Chapter 8: How to compare two treatments. Chapter

More information

DISTRIBUTIONS USED IN STATISTICAL WORK

DISTRIBUTIONS USED IN STATISTICAL WORK DISTRIBUTIONS USED IN STATISTICAL WORK In one of the classic introductory statistics books used in Education and Psychology (Glass and Stanley, 1970, Prentice-Hall) there was an excellent chapter on different

More information

Correlation & Simple Regression

Correlation & Simple Regression Chapter 11 Correlation & Simple Regression The previous chapter dealt with inference for two categorical variables. In this chapter, we would like to examine the relationship between two quantitative variables.

More information

This gives us an upper and lower bound that capture our population mean.

This gives us an upper and lower bound that capture our population mean. Confidence Intervals Critical Values Practice Problems 1 Estimation 1.1 Confidence Intervals Definition 1.1 Margin of error. The margin of error of a distribution is the amount of error we predict when

More information

21.0 Two-Factor Designs

21.0 Two-Factor Designs 21.0 Two-Factor Designs Answer Questions 1 RCBD Concrete Example Two-Way ANOVA Popcorn Example 21.4 RCBD 2 The Randomized Complete Block Design is also known as the two-way ANOVA without interaction. A

More information

The Model Building Process Part I: Checking Model Assumptions Best Practice

The Model Building Process Part I: Checking Model Assumptions Best Practice The Model Building Process Part I: Checking Model Assumptions Best Practice Authored by: Sarah Burke, PhD 31 July 2017 The goal of the STAT T&E COE is to assist in developing rigorous, defensible test

More information

The entire data set consists of n = 32 widgets, 8 of which were made from each of q = 4 different materials.

The entire data set consists of n = 32 widgets, 8 of which were made from each of q = 4 different materials. One-Way ANOVA Summary The One-Way ANOVA procedure is designed to construct a statistical model describing the impact of a single categorical factor X on a dependent variable Y. Tests are run to determine

More information

One-Way ANOVA. Some examples of when ANOVA would be appropriate include:

One-Way ANOVA. Some examples of when ANOVA would be appropriate include: One-Way ANOVA 1. Purpose Analysis of variance (ANOVA) is used when one wishes to determine whether two or more groups (e.g., classes A, B, and C) differ on some outcome of interest (e.g., an achievement

More information

SMAM 314 Practice Final Examination Winter 2003

SMAM 314 Practice Final Examination Winter 2003 SMAM 314 Practice Final Examination Winter 2003 You may use your textbook, one page of notes and a calculator. Please hand in the notes with your exam. 1. Mark the following statements True T or False

More information

The Model Building Process Part I: Checking Model Assumptions Best Practice (Version 1.1)

The Model Building Process Part I: Checking Model Assumptions Best Practice (Version 1.1) The Model Building Process Part I: Checking Model Assumptions Best Practice (Version 1.1) Authored by: Sarah Burke, PhD Version 1: 31 July 2017 Version 1.1: 24 October 2017 The goal of the STAT T&E COE

More information

1 Introduction to One-way ANOVA

1 Introduction to One-way ANOVA Review Source: Chapter 10 - Analysis of Variance (ANOVA). Example Data Source: Example problem 10.1 (dataset: exp10-1.mtw) Link to Data: http://www.auburn.edu/~carpedm/courses/stat3610/textbookdata/minitab/

More information

Analysis of Variance

Analysis of Variance Statistical Techniques II EXST7015 Analysis of Variance 15a_ANOVA_Introduction 1 Design The simplest model for Analysis of Variance (ANOVA) is the CRD, the Completely Randomized Design This model is also

More information

Chapter 7. Inference for Distributions. Introduction to the Practice of STATISTICS SEVENTH. Moore / McCabe / Craig. Lecture Presentation Slides

Chapter 7. Inference for Distributions. Introduction to the Practice of STATISTICS SEVENTH. Moore / McCabe / Craig. Lecture Presentation Slides Chapter 7 Inference for Distributions Introduction to the Practice of STATISTICS SEVENTH EDITION Moore / McCabe / Craig Lecture Presentation Slides Chapter 7 Inference for Distributions 7.1 Inference for

More information

One-Way Analysis of Variance (ANOVA)

One-Way Analysis of Variance (ANOVA) 1 One-Way Analysis of Variance (ANOVA) One-Way Analysis of Variance (ANOVA) is a method for comparing the means of a populations. This kind of problem arises in two different settings 1. When a independent

More information

Analysis of variance

Analysis of variance Analysis of variance 1 Method If the null hypothesis is true, then the populations are the same: they are normal, and they have the same mean and the same variance. We will estimate the numerical value

More information

Chapter 16. Simple Linear Regression and Correlation

Chapter 16. Simple Linear Regression and Correlation Chapter 16 Simple Linear Regression and Correlation 16.1 Regression Analysis Our problem objective is to analyze the relationship between interval variables; regression analysis is the first tool we will

More information

Inference with Simple Regression

Inference with Simple Regression 1 Introduction Inference with Simple Regression Alan B. Gelder 06E:071, The University of Iowa 1 Moving to infinite means: In this course we have seen one-mean problems, twomean problems, and problems

More information

4:3 LEC - PLANNED COMPARISONS AND REGRESSION ANALYSES

4:3 LEC - PLANNED COMPARISONS AND REGRESSION ANALYSES 4:3 LEC - PLANNED COMPARISONS AND REGRESSION ANALYSES FOR SINGLE FACTOR BETWEEN-S DESIGNS Planned or A Priori Comparisons We previously showed various ways to test all possible pairwise comparisons for

More information

Keller: Stats for Mgmt & Econ, 7th Ed July 17, 2006

Keller: Stats for Mgmt & Econ, 7th Ed July 17, 2006 Chapter 17 Simple Linear Regression and Correlation 17.1 Regression Analysis Our problem objective is to analyze the relationship between interval variables; regression analysis is the first tool we will

More information

Sleep data, two drugs Ch13.xls

Sleep data, two drugs Ch13.xls Model Based Statistics in Biology. Part IV. The General Linear Mixed Model.. Chapter 13.3 Fixed*Random Effects (Paired t-test) ReCap. Part I (Chapters 1,2,3,4), Part II (Ch 5, 6, 7) ReCap Part III (Ch

More information

Keppel, G. & Wickens, T. D. Design and Analysis Chapter 4: Analytical Comparisons Among Treatment Means

Keppel, G. & Wickens, T. D. Design and Analysis Chapter 4: Analytical Comparisons Among Treatment Means Keppel, G. & Wickens, T. D. Design and Analysis Chapter 4: Analytical Comparisons Among Treatment Means 4.1 The Need for Analytical Comparisons...the between-groups sum of squares averages the differences

More information

Statistics for Managers using Microsoft Excel 6 th Edition

Statistics for Managers using Microsoft Excel 6 th Edition Statistics for Managers using Microsoft Excel 6 th Edition Chapter 13 Simple Linear Regression 13-1 Learning Objectives In this chapter, you learn: How to use regression analysis to predict the value of

More information

How To: Analyze a Split-Plot Design Using STATGRAPHICS Centurion

How To: Analyze a Split-Plot Design Using STATGRAPHICS Centurion How To: Analyze a SplitPlot Design Using STATGRAPHICS Centurion by Dr. Neil W. Polhemus August 13, 2005 Introduction When performing an experiment involving several factors, it is best to randomize the

More information

Analysis of Covariance. The following example illustrates a case where the covariate is affected by the treatments.

Analysis of Covariance. The following example illustrates a case where the covariate is affected by the treatments. Analysis of Covariance In some experiments, the experimental units (subjects) are nonhomogeneous or there is variation in the experimental conditions that are not due to the treatments. For example, a

More information

Chapter 16. Simple Linear Regression and dcorrelation

Chapter 16. Simple Linear Regression and dcorrelation Chapter 16 Simple Linear Regression and dcorrelation 16.1 Regression Analysis Our problem objective is to analyze the relationship between interval variables; regression analysis is the first tool we will

More information

1. Least squares with more than one predictor

1. Least squares with more than one predictor Statistics 1 Lecture ( November ) c David Pollard Page 1 Read M&M Chapter (skip part on logistic regression, pages 730 731). Read M&M pages 1, for ANOVA tables. Multiple regression. 1. Least squares with

More information

In ANOVA the response variable is numerical and the explanatory variables are categorical.

In ANOVA the response variable is numerical and the explanatory variables are categorical. 1 ANOVA ANOVA means ANalysis Of VAriance. The ANOVA is a tool for studying the influence of one or more qualitative variables on the mean of a numerical variable in a population. In ANOVA the response

More information

Ch 13 & 14 - Regression Analysis

Ch 13 & 14 - Regression Analysis Ch 3 & 4 - Regression Analysis Simple Regression Model I. Multiple Choice:. A simple regression is a regression model that contains a. only one independent variable b. only one dependent variable c. more

More information

Review of Statistics 101

Review of Statistics 101 Review of Statistics 101 We review some important themes from the course 1. Introduction Statistics- Set of methods for collecting/analyzing data (the art and science of learning from data). Provides methods

More information

13: Additional ANOVA Topics. Post hoc Comparisons

13: Additional ANOVA Topics. Post hoc Comparisons 13: Additional ANOVA Topics Post hoc Comparisons ANOVA Assumptions Assessing Group Variances When Distributional Assumptions are Severely Violated Post hoc Comparisons In the prior chapter we used ANOVA

More information

Inference for Regression Inference about the Regression Model and Using the Regression Line

Inference for Regression Inference about the Regression Model and Using the Regression Line Inference for Regression Inference about the Regression Model and Using the Regression Line PBS Chapter 10.1 and 10.2 2009 W.H. Freeman and Company Objectives (PBS Chapter 10.1 and 10.2) Inference about

More information

CHAPTER 13: F PROBABILITY DISTRIBUTION

CHAPTER 13: F PROBABILITY DISTRIBUTION CHAPTER 13: F PROBABILITY DISTRIBUTION continuous probability distribution skewed to the right variable values on horizontal axis are 0 area under the curve represents probability horizontal asymptote

More information

CHAPTER 1. Introduction

CHAPTER 1. Introduction CHAPTER 1 Introduction Engineers and scientists are constantly exposed to collections of facts, or data. The discipline of statistics provides methods for organizing and summarizing data, and for drawing

More information

What Is ANOVA? Comparing Groups. One-way ANOVA. One way ANOVA (the F ratio test)

What Is ANOVA? Comparing Groups. One-way ANOVA. One way ANOVA (the F ratio test) What Is ANOVA? One-way ANOVA ANOVA ANalysis Of VAriance ANOVA compares the means of several groups. The groups are sometimes called "treatments" First textbook presentation in 95. Group Group σ µ µ σ µ

More information

CHAPTER 10 ONE-WAY ANALYSIS OF VARIANCE. It would be very unusual for all the research one might conduct to be restricted to

CHAPTER 10 ONE-WAY ANALYSIS OF VARIANCE. It would be very unusual for all the research one might conduct to be restricted to CHAPTER 10 ONE-WAY ANALYSIS OF VARIANCE It would be very unusual for all the research one might conduct to be restricted to comparisons of only two samples. Respondents and various groups are seldom divided

More information

1-Way ANOVA MATH 143. Spring Department of Mathematics and Statistics Calvin College

1-Way ANOVA MATH 143. Spring Department of Mathematics and Statistics Calvin College 1-Way ANOVA MATH 143 Department of Mathematics and Statistics Calvin College Spring 2010 The basic ANOVA situation Two variables: 1 Categorical, 1 Quantitative Main Question: Do the (means of) the quantitative

More information

Ph.D. Preliminary Examination Statistics June 2, 2014

Ph.D. Preliminary Examination Statistics June 2, 2014 Ph.D. Preliminary Examination Statistics June, 04 NOTES:. The exam is worth 00 points.. Partial credit may be given for partial answers if possible.. There are 5 pages in this exam paper. I have neither

More information

Inferential statistics

Inferential statistics Inferential statistics Inference involves making a Generalization about a larger group of individuals on the basis of a subset or sample. Ahmed-Refat-ZU Null and alternative hypotheses In hypotheses testing,

More information

Basic Business Statistics 6 th Edition

Basic Business Statistics 6 th Edition Basic Business Statistics 6 th Edition Chapter 12 Simple Linear Regression Learning Objectives In this chapter, you learn: How to use regression analysis to predict the value of a dependent variable based

More information

Chapter 1 Statistical Inference

Chapter 1 Statistical Inference Chapter 1 Statistical Inference causal inference To infer causality, you need a randomized experiment (or a huge observational study and lots of outside information). inference to populations Generalizations

More information

CHAPTER 13: F PROBABILITY DISTRIBUTION

CHAPTER 13: F PROBABILITY DISTRIBUTION CHAPTER 13: F PROBABILITY DISTRIBUTION continuous probability distribution skewed to the right variable values on horizontal axis are 0 area under the curve represents probability horizontal asymptote

More information

Two-Way Analysis of Variance - no interaction

Two-Way Analysis of Variance - no interaction 1 Two-Way Analysis of Variance - no interaction Example: Tests were conducted to assess the effects of two factors, engine type, and propellant type, on propellant burn rate in fired missiles. Three engine

More information

MULTIPLE LINEAR REGRESSION IN MINITAB

MULTIPLE LINEAR REGRESSION IN MINITAB MULTIPLE LINEAR REGRESSION IN MINITAB This document shows a complicated Minitab multiple regression. It includes descriptions of the Minitab commands, and the Minitab output is heavily annotated. Comments

More information

W&M CSCI 688: Design of Experiments Homework 2. Megan Rose Bryant

W&M CSCI 688: Design of Experiments Homework 2. Megan Rose Bryant W&M CSCI 688: Design of Experiments Homework 2 Megan Rose Bryant September 25, 201 3.5 The tensile strength of Portland cement is being studied. Four different mixing techniques can be used economically.

More information

An Analysis of College Algebra Exam Scores December 14, James D Jones Math Section 01

An Analysis of College Algebra Exam Scores December 14, James D Jones Math Section 01 An Analysis of College Algebra Exam s December, 000 James D Jones Math - Section 0 An Analysis of College Algebra Exam s Introduction Students often complain about a test being too difficult. Are there

More information

Lec 1: An Introduction to ANOVA

Lec 1: An Introduction to ANOVA Ying Li Stockholm University October 31, 2011 Three end-aisle displays Which is the best? Design of the Experiment Identify the stores of the similar size and type. The displays are randomly assigned to

More information

Warm-up Using the given data Create a scatterplot Find the regression line

Warm-up Using the given data Create a scatterplot Find the regression line Time at the lunch table Caloric intake 21.4 472 30.8 498 37.7 335 32.8 423 39.5 437 22.8 508 34.1 431 33.9 479 43.8 454 42.4 450 43.1 410 29.2 504 31.3 437 28.6 489 32.9 436 30.6 480 35.1 439 33.0 444

More information

Multiple Regression Examples

Multiple Regression Examples Multiple Regression Examples Example: Tree data. we have seen that a simple linear regression of usable volume on diameter at chest height is not suitable, but that a quadratic model y = β 0 + β 1 x +

More information

INFERENCE FOR REGRESSION

INFERENCE FOR REGRESSION CHAPTER 3 INFERENCE FOR REGRESSION OVERVIEW In Chapter 5 of the textbook, we first encountered regression. The assumptions that describe the regression model we use in this chapter are the following. We

More information

2.4.3 Estimatingσ Coefficient of Determination 2.4. ASSESSING THE MODEL 23

2.4.3 Estimatingσ Coefficient of Determination 2.4. ASSESSING THE MODEL 23 2.4. ASSESSING THE MODEL 23 2.4.3 Estimatingσ 2 Note that the sums of squares are functions of the conditional random variables Y i = (Y X = x i ). Hence, the sums of squares are random variables as well.

More information

CHAPTER 5 FUNCTIONAL FORMS OF REGRESSION MODELS

CHAPTER 5 FUNCTIONAL FORMS OF REGRESSION MODELS CHAPTER 5 FUNCTIONAL FORMS OF REGRESSION MODELS QUESTIONS 5.1. (a) In a log-log model the dependent and all explanatory variables are in the logarithmic form. (b) In the log-lin model the dependent variable

More information

ANALYSIS OF VARIANCE OF BALANCED DAIRY SCIENCE DATA USING SAS

ANALYSIS OF VARIANCE OF BALANCED DAIRY SCIENCE DATA USING SAS ANALYSIS OF VARIANCE OF BALANCED DAIRY SCIENCE DATA USING SAS Ravinder Malhotra and Vipul Sharma National Dairy Research Institute, Karnal-132001 The most common use of statistics in dairy science is testing

More information

PSYC 331 STATISTICS FOR PSYCHOLOGISTS

PSYC 331 STATISTICS FOR PSYCHOLOGISTS PSYC 331 STATISTICS FOR PSYCHOLOGISTS Session 4 A PARAMETRIC STATISTICAL TEST FOR MORE THAN TWO POPULATIONS Lecturer: Dr. Paul Narh Doku, Dept of Psychology, UG Contact Information: pndoku@ug.edu.gh College

More information

Chap The McGraw-Hill Companies, Inc. All rights reserved.

Chap The McGraw-Hill Companies, Inc. All rights reserved. 11 pter11 Chap Analysis of Variance Overview of ANOVA Multiple Comparisons Tests for Homogeneity of Variances Two-Factor ANOVA Without Replication General Linear Model Experimental Design: An Overview

More information

An Old Research Question

An Old Research Question ANOVA An Old Research Question The impact of TV on high-school grade Watch or not watch Two groups The impact of TV hours on high-school grade Exactly how much TV watching would make difference Multiple

More information

Chapter 14 Multiple Regression Analysis

Chapter 14 Multiple Regression Analysis Chapter 14 Multiple Regression Analysis 1. a. Multiple regression equation b. the Y-intercept c. $374,748 found by Y ˆ = 64,1 +.394(796,) + 9.6(694) 11,6(6.) (LO 1) 2. a. Multiple regression equation b.

More information

MULTIPLE REGRESSION METHODS

MULTIPLE REGRESSION METHODS DEPARTMENT OF POLITICAL SCIENCE AND INTERNATIONAL RELATIONS Posc/Uapp 816 MULTIPLE REGRESSION METHODS I. AGENDA: A. Residuals B. Transformations 1. A useful procedure for making transformations C. Reading:

More information

Analysis of 2x2 Cross-Over Designs using T-Tests

Analysis of 2x2 Cross-Over Designs using T-Tests Chapter 234 Analysis of 2x2 Cross-Over Designs using T-Tests Introduction This procedure analyzes data from a two-treatment, two-period (2x2) cross-over design. The response is assumed to be a continuous

More information

Factorial designs. Experiments

Factorial designs. Experiments Chapter 5: Factorial designs Petter Mostad mostad@chalmers.se Experiments Actively making changes and observing the result, to find causal relationships. Many types of experimental plans Measuring response

More information

Your schedule of coming weeks. One-way ANOVA, II. Review from last time. Review from last time /22/2004. Create ANOVA table

Your schedule of coming weeks. One-way ANOVA, II. Review from last time. Review from last time /22/2004. Create ANOVA table Your schedule of coming weeks One-way ANOVA, II 9.07 //00 Today: One-way ANOVA, part II Next week: Two-way ANOVA, parts I and II. One-way ANOVA HW due Thursday Week of May Teacher out of town all week

More information

, (1) e i = ˆσ 1 h ii. c 2016, Jeffrey S. Simonoff 1

, (1) e i = ˆσ 1 h ii. c 2016, Jeffrey S. Simonoff 1 Regression diagnostics As is true of all statistical methodologies, linear regression analysis can be a very effective way to model data, as along as the assumptions being made are true. For the regression

More information

Describing distributions with numbers

Describing distributions with numbers Describing distributions with numbers A large number or numerical methods are available for describing quantitative data sets. Most of these methods measure one of two data characteristics: The central

More information

Factorial Analysis of Variance

Factorial Analysis of Variance Factorial Analysis of Variance Conceptual Example A repeated-measures t-test is more likely to lead to rejection of the null hypothesis if a) *Subjects show considerable variability in their change scores.

More information

psyc3010 lecture 2 factorial between-ps ANOVA I: omnibus tests

psyc3010 lecture 2 factorial between-ps ANOVA I: omnibus tests psyc3010 lecture 2 factorial between-ps ANOVA I: omnibus tests last lecture: introduction to factorial designs next lecture: factorial between-ps ANOVA II: (effect sizes and follow-up tests) 1 general

More information

1 One-way Analysis of Variance

1 One-way Analysis of Variance 1 One-way Analysis of Variance Suppose that a random sample of q individuals receives treatment T i, i = 1,,... p. Let Y ij be the response from the jth individual to be treated with the ith treatment

More information

Mathematical Notation Math Introduction to Applied Statistics

Mathematical Notation Math Introduction to Applied Statistics Mathematical Notation Math 113 - Introduction to Applied Statistics Name : Use Word or WordPerfect to recreate the following documents. Each article is worth 10 points and should be emailed to the instructor

More information

EXPERIMENT 2 Reaction Time Objectives Theory

EXPERIMENT 2 Reaction Time Objectives Theory EXPERIMENT Reaction Time Objectives to make a series of measurements of your reaction time to make a histogram, or distribution curve, of your measured reaction times to calculate the "average" or mean

More information

RCB - Example. STA305 week 10 1

RCB - Example. STA305 week 10 1 RCB - Example An accounting firm wants to select training program for its auditors who conduct statistical sampling as part of their job. Three training methods are under consideration: home study, presentations

More information

INTRODUCTION TO DESIGN AND ANALYSIS OF EXPERIMENTS

INTRODUCTION TO DESIGN AND ANALYSIS OF EXPERIMENTS GEORGE W. COBB Mount Holyoke College INTRODUCTION TO DESIGN AND ANALYSIS OF EXPERIMENTS Springer CONTENTS To the Instructor Sample Exam Questions To the Student Acknowledgments xv xxi xxvii xxix 1. INTRODUCTION

More information

(1) The explanatory or predictor variables may be qualitative. (We ll focus on examples where this is the case.)

(1) The explanatory or predictor variables may be qualitative. (We ll focus on examples where this is the case.) Introduction to Analysis of Variance Analysis of variance models are similar to regression models, in that we re interested in learning about the relationship between a dependent variable (a response)

More information

Postgraduate Certificate in Statistics. Design and Analysis of Experiments Module

Postgraduate Certificate in Statistics. Design and Analysis of Experiments Module Chapter Postgraduate Certificate in Statistics Design and Analysis of Experiments Basic Definitions, Design Principles and Designs.1 Introduction... 1. Definitions....3 Simple comparative experiments...

More information

Simple Linear Regression: One Quantitative IV

Simple Linear Regression: One Quantitative IV Simple Linear Regression: One Quantitative IV Linear regression is frequently used to explain variation observed in a dependent variable (DV) with theoretically linked independent variables (IV). For example,

More information

Design & Analysis of Experiments 7E 2009 Montgomery

Design & Analysis of Experiments 7E 2009 Montgomery 1 What If There Are More Than Two Factor Levels? The t-test does not directly apply ppy There are lots of practical situations where there are either more than two levels of interest, or there are several

More information

DESAIN EKSPERIMEN Analysis of Variances (ANOVA) Semester Genap 2017/2018 Jurusan Teknik Industri Universitas Brawijaya

DESAIN EKSPERIMEN Analysis of Variances (ANOVA) Semester Genap 2017/2018 Jurusan Teknik Industri Universitas Brawijaya DESAIN EKSPERIMEN Analysis of Variances (ANOVA) Semester Jurusan Teknik Industri Universitas Brawijaya Outline Introduction The Analysis of Variance Models for the Data Post-ANOVA Comparison of Means Sample

More information

Multiple comparisons - subsequent inferences for two-way ANOVA

Multiple comparisons - subsequent inferences for two-way ANOVA 1 Multiple comparisons - subsequent inferences for two-way ANOVA the kinds of inferences to be made after the F tests of a two-way ANOVA depend on the results if none of the F tests lead to rejection of

More information

In a one-way ANOVA, the total sums of squares among observations is partitioned into two components: Sums of squares represent:

In a one-way ANOVA, the total sums of squares among observations is partitioned into two components: Sums of squares represent: Activity #10: AxS ANOVA (Repeated subjects design) Resources: optimism.sav So far in MATH 300 and 301, we have studied the following hypothesis testing procedures: 1) Binomial test, sign-test, Fisher s

More information

Unit 27 One-Way Analysis of Variance

Unit 27 One-Way Analysis of Variance Unit 27 One-Way Analysis of Variance Objectives: To perform the hypothesis test in a one-way analysis of variance for comparing more than two population means Recall that a two sample t test is applied

More information

Business 320, Fall 1999, Final

Business 320, Fall 1999, Final Business 320, Fall 1999, Final name You may use a calculator and two cheat sheets. You have 3 hours. I pledge my honor that I have not violated the Honor Code during this examination. Obvioiusly, you may

More information

Inferences for Regression

Inferences for Regression Inferences for Regression An Example: Body Fat and Waist Size Looking at the relationship between % body fat and waist size (in inches). Here is a scatterplot of our data set: Remembering Regression In

More information

In the previous chapter, we learned how to use the method of least-squares

In the previous chapter, we learned how to use the method of least-squares 03-Kahane-45364.qxd 11/9/2007 4:40 PM Page 37 3 Model Performance and Evaluation In the previous chapter, we learned how to use the method of least-squares to find a line that best fits a scatter of points.

More information

Glossary. The ISI glossary of statistical terms provides definitions in a number of different languages:

Glossary. The ISI glossary of statistical terms provides definitions in a number of different languages: Glossary The ISI glossary of statistical terms provides definitions in a number of different languages: http://isi.cbs.nl/glossary/index.htm Adjusted r 2 Adjusted R squared measures the proportion of the

More information

Chapter 5 Introduction to Factorial Designs Solutions

Chapter 5 Introduction to Factorial Designs Solutions Solutions from Montgomery, D. C. (1) Design and Analysis of Experiments, Wiley, NY Chapter 5 Introduction to Factorial Designs Solutions 5.1. The following output was obtained from a computer program that

More information

Confidence Intervals. Confidence interval for sample mean. Confidence interval for sample mean. Confidence interval for sample mean

Confidence Intervals. Confidence interval for sample mean. Confidence interval for sample mean. Confidence interval for sample mean Confidence Intervals Confidence interval for sample mean The CLT tells us: as the sample size n increases, the sample mean is approximately Normal with mean and standard deviation Thus, we have a standard

More information

Residual Analysis for two-way ANOVA The twoway model with K replicates, including interaction,

Residual Analysis for two-way ANOVA The twoway model with K replicates, including interaction, Residual Analysis for two-way ANOVA The twoway model with K replicates, including interaction, is Y ijk = µ ij + ɛ ijk = µ + α i + β j + γ ij + ɛ ijk with i = 1,..., I, j = 1,..., J, k = 1,..., K. In carrying

More information

Design of Engineering Experiments Part 2 Basic Statistical Concepts Simple comparative experiments

Design of Engineering Experiments Part 2 Basic Statistical Concepts Simple comparative experiments Design of Engineering Experiments Part 2 Basic Statistical Concepts Simple comparative experiments The hypothesis testing framework The two-sample t-test Checking assumptions, validity Comparing more that

More information

Orthogonal contrasts for a 2x2 factorial design Example p130

Orthogonal contrasts for a 2x2 factorial design Example p130 Week 9: Orthogonal comparisons for a 2x2 factorial design. The general two-factor factorial arrangement. Interaction and additivity. ANOVA summary table, tests, CIs. Planned/post-hoc comparisons for the

More information

Multiple Comparison Methods for Means

Multiple Comparison Methods for Means SIAM REVIEW Vol. 44, No. 2, pp. 259 278 c 2002 Society for Industrial and Applied Mathematics Multiple Comparison Methods for Means John A. Rafter Martha L. Abell James P. Braselton Abstract. Multiple

More information

Notes on Maxwell & Delaney

Notes on Maxwell & Delaney Notes on Maxwell & Delaney PSY710 9 Designs with Covariates 9.1 Blocking Consider the following hypothetical experiment. We want to measure the effect of a drug on locomotor activity in hyperactive children.

More information

Review. One-way ANOVA, I. What s coming up. Multiple comparisons

Review. One-way ANOVA, I. What s coming up. Multiple comparisons Review One-way ANOVA, I 9.07 /15/00 Earlier in this class, we talked about twosample z- and t-tests for the difference between two conditions of an independent variable Does a trial drug work better than

More information

The simple linear regression model discussed in Chapter 13 was written as

The simple linear regression model discussed in Chapter 13 was written as 1519T_c14 03/27/2006 07:28 AM Page 614 Chapter Jose Luis Pelaez Inc/Blend Images/Getty Images, Inc./Getty Images, Inc. 14 Multiple Regression 14.1 Multiple Regression Analysis 14.2 Assumptions of the Multiple

More information