World Library  
Flag as Inappropriate
Email this Article

Dunnett's test

Article Id: WHEBN0036510069
Reproduction Date:

Title: Dunnett's test  
Author: World Heritage Encyclopedia
Language: English
Subject: Multiple comparisons, Statistical tests, Post hoc analysis, Multiple comparisons problem, Analysis of variance
Collection: Multiple Comparisons, Statistical Tests
Publisher: World Heritage Encyclopedia
Publication
Date:
 

Dunnett's test

In statistics, Dunnett's test is a multiple comparison procedure[1] developed by Canadian statistician Charles Dunnett[2] to compare each of a number of treatments with a single control.[3][4] Multiple comparisons to a control are also referred to as many-to-one comparisons.

Contents

  • History 1
    • Multiple Comparisons Problem 1.1
    • Uses of Dunnett’s test 1.2
  • Formal description of Dunnett's test 2
    • Assumptions 2.1
    • Calculation 2.2
  • Examples 3
    • Breaking strength of fabric 3.1
  • References 4

History

Dunnett's test was developed in 1955;[5] an updated table of critical values was published in 1964.[6]

Multiple Comparisons Problem

The multiple comparisons, multiplicity or multiple testing problem occurs when one considers a set of statistical inferences simultaneously or infers a subset of parameters selected based on the observed values. The major issue in any discussion of multiple-comparison procedures is the question of the probability of Type I errors. Most differences among alternative techniques result from different approaches to the question of how to control these errors. The problem is in part technical; but it is really much more a subjective question of how you want to define the error rate and how large you are willing to let the maximum possible error rate be.[7] Dunnett's test are well known and widely used in multiple comparison procedure for simultaneously comparing, by interval estimation or hypothesis testing, all active treatments with a control when sampling from a distribution where the normality assumption is reasonable. Dunnett's test is designed to hold the familywise error rate at or below \alpha when performing multiple comparisons of treatment group with control.[7]

Uses of Dunnett’s test

The original work on Multiple Comparisons problem was made by Tukey and Scheffé. Their method was a general one, which considered all kinds of pairwise comparisons.[7] Tukey's and Scheffé's methods allow any number of comparisons among a set of sample means. On the other hand, Dunnett's test only compares one group with the others, addressing a special case of multiple comparisons problem — pairwise comparisons of a treatment group with control groups. In the general case, where we compare each of the pairs, we make \frac{k(k-1)}{2} comparisons (where k is the number of groups), but in the treatment vs. controls case we will make only (k-1) comparisons. If in the case of treatment and control groups we were to use the more general Tukey's and Scheffé's methods, it can result in unnecessary wider confidence intervals. Dunnett's test takes into consideration the special structure of comparing treatment against control and results in narrower confidence intervals.[5]
It is very common to use Dunnett's test in medical experiments, for example comparing blood count measurements on three groups of animals, one of which served as a control while the other two were treated with two different drugs. Another common use of this method is among agronomists: agronomists may want to study the effect of certain chemicals added to the soil on crop yield, so they will leave some plots untreated (control plots) and compare them to the plots where chemicals were added to the soil (treatment plots).

Formal description of Dunnett's test

Dunnett's test is performed by computing a Student's t-statistic for each experimental, or treatment, group where the statistic compares the treatment group to a single control group.[8][9] Since each comparison has the same control in common, the procedure incorporates the dependencies between these comparisons. In particular, the t-statistics are all derived from the same estimate of the error variance which is obtained by pooling the sums of squares for error across all (treatment and control) groups. The formal test statistic for Dunnett's test is either the largest in absolute value of these t-statistics (if a two-tailed test is required), or the most negative or most positive of the t-statistics (if a one-tailed test is required).

In Dunnett's test we can use a common table of critical values, but more flexible options are nowadays readily available in many statistics packages such as R. The critical values for any given percentage point depend on: whether a one- or- two-tailed test is performed; the number of groups being compared; the overall number of trials.

Assumptions

The analysis considers the case where the results of the experiment are numerical, and the experiment is performed to compare p treatments with a control group. The results can be summarized as a set of (p+1) calculated means of the sets of observations, (\bar{X_{0}},...,\bar{X_{p}}), while (\bar{X_{1}},...,\bar{X_{p}}) are referring to the treatment and \bar{X_{0}} is referring to the control set of observations, and s is an independent estimate of the common standard deviation of all p+1 sets of observations. All \bar{X_{i}} of the p+1 sets of observations are assumed to be independently and normally distributed with a common variance \sigma^2 and means \mu_{i}. There is also an assumption that there is an available estimate s^2 for \sigma^2.

Calculation

Calculation Dunnett's test's calculation is a procedure that is based on calculating confidence statements about the true or the expected values of the p differences \bar{X_{i}}-\bar{X_{0}}, thus the differences between treatment groups' mean and control group's mean. This procedure enables that the probability of all p statements \bar{X_{i}}-\bar{X_{0}} being simultaneously correct is equal to a specified value,P. When calculating one sided upper (or lower) Confidence interval for the true value of the difference between the mean of the treatment and the control group, P constitutes the probability that this actual value will be less than the upper (or greater than the lower) limit of that interval. When calculating two-sided confidence interval, P constitutes the probability that the true value will be between the upper and the lower limits.

First, we will denote the available N observations by X_{ij} when i=1...p and j=1...N_{i} and estimate the common variance by, for example: s^2= \frac{\sum_{i=0}^{p}\sum_{j=1}^{N_{i}} (X_{ij}-\bar{X_{i}})}{n} when \bar{X_{i}} is the mean of group i and N_{i} is the number of observations in group i, and n=\sum_{i=0}^{p} N_{i}-(p+1) degrees of freedom. As mentioned before, we would like to obtain separate confidence limits for each of the differences m_{i}-m_{0}, (i=1...p) such that the probability that all p confidence intervals will contain the corresponding m_{i}-m_{0} is equal to P.

We will consider the general case where there are p treatment groups and one control group. We will write:

z_{i}= \cfrac{\bar{X_{i}}-\bar{X_{0}}-(m_{i}-m_{0})}{\sqrt{ \cfrac{1}{N_{i}}+ \cfrac{1}{N_{0}}}}

D_{i}= \cfrac{\bar{X_{i}}-\bar{X_{0}}-(m_{i}-m_{0})}{s\sqrt{ \cfrac{1}{N_{i}}+ \cfrac{1}{N_{0}}}}

we will also write: D_{i}=\frac{z_{i}}{s}, which follows the Student's t-statistic distribution with n degrees of freedom. The lower confidence limits with joint confidence coefficient P for the p treatment effects m_{i}-m_{0}, (i=1...p) will be given by:

\bar{X_{i}}-\bar{X_{0}}-d_{i}'s\sqrt{ \frac{1}{N_{i}}+\frac{1}{N_{0}}} , i=1...p

and the p constants d_{i}' are chosen so that Prob(t_{1} Similarly, the upper limits will be given by:

\bar{X_{i}}-\bar{X_{0}}+d_{i}'s\sqrt{ \frac{1}{N_{i}}+\frac{1}{N_{0}}} , i=1...p

For bounding m_{i}-m_{0} in both directions, the following interval might be taken:

\bar{X_{i}}-\bar{X_{0}}\pm d_{i}'s\sqrt{ \frac{1}{N_{i}}+\frac{1}{N_{0}}} , i=1...p

when d_{i}'' are chosen to satisfy Prob(|t_{1}|. The solution to those particular values of d_{i}'' for two sided test and d_{i}' for one sided test is given in the tables.[5] An updated table of critical values was published in 1964.[6]

Examples

Breaking strength of fabric[5]

The following example was adapted from one given by Villars[6]. The data represent measurements on the breaking strength of fabric treated by three different chemical process compared with a standard method of manufacture.

breaking strength(lbs.)
standard process 1 process 2 process 3
55 55 55 50
47 64 49 44
48 64 52 41
Means 50 61 52 45
Variance 19 27 9 21

Here , p=3 and N=3. The average variance is s^2=19, which is an estimate of the common variance of the four sets with (p+1)(N-1)=8 degrees of freedom. This can be calculated as follows:

\frac {55^2+47^2+48^2+55^2+...+41^2-3(50^2+61^2+52^2+45^2)}{8}=\frac{152}{8}=19.

The standard deviation is s=\sqrt{19}=4.36 and the estimated standard error of a difference between two means is s\sqrt{\frac{2}{N}}=4.36\sqrt{\frac{2}{N}}=3.56 .

The quantity which must be added to and/or subtracted from the observed differences between the means to give their confidence limits has been called by Tukey an "allowance" and is given by A=t_s\sqrt{\frac{2}{N}}, where t is obtained from Dunnett's Table 1 if one side limits are desired or from Dunnett's Table 2 if two-sided limits are wanted. For p=3 and d.f.=8 , t=2.42 for one side limits and t=2.88 for two-sided limits for p=95%. Analogous values of t can be determined from the tables if p=99% confidence is required. For one-sided limits, the allowance is A=(2.42)(3.56)=9 and the experimenter can conclude that:

  • The breaking strength using process 1 exceeds the standard by at least 61-50-9=2lbs.
  • The breaking strength using process 2 exceeds the standard by at least 52-50-9=-7lbs.
  • The breaking strength using process 3 exceeds the standard by at least 45-50-9=-14lbs.

The joint statment consisting of the above three conclusions has a confidence coefficient of 95%, i.e., in the long run, 95% of such joint statments will actually be correct. Upper limits for the three differences could be obtained in an analogous manner. For two-sided limits, the allowance is A=(2.94)(3.56)=11 and the experimenter can conclude that:

  • The breaking strength using process 1 exceeds the standard by an amount between

61-50-11=0lbs. and 61-50+11=22lbs.

  • The breaking strength using process 2 exceeds the standard by an amount between

52-50-11=-9lbs and 52-50+11=13lbs.

  • The breaking strength using process 3 exceeds the standard by an amount between

45-50-11=-16lbs and 45-50+11=6lbs. The joint confidence coefficient for these three statement is greater then 95%. (Due to an approximation made in computing Tables 2a and 2b, the tabulated values of t are somewhat larger then necessary so that the actual p's attained are slightly greater than 95 and 99%.No such approximation was made in computing Tables 1a and 1b).

References

  1. ^ Upton G. & Cook I. (2006.) A Dictionary of Statistics, 2e, Oxford University Press, Oxford, United Kingdom.
  2. ^ Statistics II for Dummies - Deborah Rumsey - Google Books. Books.google.com. 2009-08-19. Retrieved 2012-08-22. 
  3. ^ Everett B. S. & Shrondal A. (2010.) The Cambridge Dictionary of Statistics, 4e, Cambridge University Press, Cambridge, United Kingdom.
  4. ^ "Statistical Software | University of Kentucky Information Technology". Uky.edu. Retrieved 2012-08-22. 
  5. ^ a b c d Dunnett C. W. (1955.) "A multiple comparison procedure for comparing several treatments with a control", Journal of the American Statistical Association, 50:1096–1121.
  6. ^ a b Dunnett C. W. (1964.) "New tables for multiple comparisons with a control", Biometrics, 20:482–491.
  7. ^ a b c David C. Howell, "Statistical Methods for Psychology",8th ed.
  8. ^ Dunnett's test, HyperStat Online: An Introductory Statistics Textbook and Online Tutorial for Help in Statistics Courses
  9. ^ Mechanics of Different Tests - Biostatistics BI 345, Saint Anselm College
This article was sourced from Creative Commons Attribution-ShareAlike License; additional terms may apply. World Heritage Encyclopedia content is assembled from numerous content providers, Open Access Publishing, and in compliance with The Fair Access to Science and Technology Research Act (FASTR), Wikimedia Foundation, Inc., Public Library of Science, The Encyclopedia of Life, Open Book Publishers (OBP), PubMed, U.S. National Library of Medicine, National Center for Biotechnology Information, U.S. National Library of Medicine, National Institutes of Health (NIH), U.S. Department of Health & Human Services, and USA.gov, which sources content from all federal, state, local, tribal, and territorial government publication portals (.gov, .mil, .edu). Funding for USA.gov and content contributors is made possible from the U.S. Congress, E-Government Act of 2002.
 
Crowd sourced content that is contributed to World Heritage Encyclopedia is peer reviewed and edited by our editorial staff to ensure quality scholarly research articles.
 
By using this site, you agree to the Terms of Use and Privacy Policy. World Heritage Encyclopedia™ is a registered trademark of the World Public Library Association, a non-profit organization.
 


Copyright © World Library Foundation. All rights reserved. eBooks from Project Gutenberg are sponsored by the World Library Foundation,
a 501c(4) Member's Support Non-Profit Organization, and is NOT affiliated with any governmental agency or department.