Repeated Measures Design

11 Key excerpts on "Repeated Measures Design"

• 

  eBook - PDF

  Modern Experimental Design

  • Thomas P. Ryan(Author)
  • 2006(Publication Date)
  • Wiley-Interscience

    (Publisher)

  CHAPTER 11 Repeated Measures Designs Repeated Measures Designs are designs for which, as the name suggests, repeated measurements are made on the experimental units, which are typically people as the design is often used in clinical work and in education. Certainly, repeated measures on the same subject are a necessary part of any design for studying rates of learning as a function of treatment effects. In clinical work, subjects may be measured over time after receiving some type of treatment. At the outset, we should note that not much attention is given to repeated mea-sures designs in most books on experimental design; exceptions are Hinkelmann and Kempthorne (2005), Giesbrecht and Gumpertz (2004), Kuehl (2000), and Winer, Brown, and Michels (1991). In particular, the latter has one chapter on single-factor designs with repeated measures and a chapter on multifactor designs with repeated measures. The best book that is devoted almost exclusively to repeated measures de-signs is probably Davis (2002). Other books on the subject include Lindsey (1999), Crowder and Hand (1990), Vonesh and Chinchilli (1997), Girden (1992), and David-ian and Giltinan (1995). Although somewhat outdated, Fleiss (1986) has two chap-ters on Repeated Measures Designs. Review papers on the subject include Keselman, Algina, and Kowalchuk (2001), which is on the analysis of data from repeated mea-sures designs, as is Everitt (1995), who provides illustrative examples. There are both advantages and disadvantages associated with these designs. One advantage is that information on the time trend of the response variable is available under different treatment conditions. Furthermore, each subject serves as his or her own control, which permits the use of a smaller sample size. More specifically, subject-to-subject variation is not problematic when each subject receives all the treatments instead of each subject receiving only one treatment.

  Sign up to read

  Learn more about book
• 

  eBook - ePub

  Repeated Measures Design for Empirical Researchers

  • J. P. Verma(Author)
  • 2015(Publication Date)
  • Wiley

    (Publisher)

  within-subjects design.

  Advantage of One-Way Repeated Measures Design

  This design has the following benefits:

  1. 1. Less number of subjects is required for the study due to which an experimenter can have more control in the experiment.
  2. 2. Since same subjects are tested in all treatments, subjects serve their own control due to which experimental error is reduced.
  3. 3. The Repeated Measures Design is more efficient in comparison to that of the independent measures design because a part of the treatment group's variability is explained by the subjects thereby experimental error is reduced.
  4. 4. This design is very sensitive in nature due to which a slight change in the criterion variable due to the manipulation of treatment conditions can be easily detected.
  5. 5. This design allows the experimenter to study the behavior of subjects due to intervention of treatment over the period of time.

  Weakness of Repeated Measures Design

  Due to carryover effect performance of the subjects may be affected in different treatment conditions. In other words, participation in one treatment may affect the performance of subjects in other treatment groups resulting creation of a confounding extraneous variable that varies with the independent variable. The carryover effect may be because of the fatigue experienced or due to learning in earlier treatment conditions. Another weakness of this design is that if a subject is tested in all treatments in a specific sequence an order effect may be generated which affects the performance of the subjects. Further, if the number of treatments are large, subjects may loose interest in the experiment as they might feel bore which in turn affects the outcomes in the study.

  Sign up to read

  Learn more about book
• 

  eBook - ePub

  Statistics for Psychologists

  An Intermediate Course

  • Brian S. Everitt(Author)
  • 2001(Publication Date)
  • Taylor & Francis

    (Publisher)

  5 Analysis of Repeated Measure Designs 5.1.  Introduction

  Many studies undertaken in the behavioral sciences and related disciplines involve recording the value of a response variable for each subject under more than one condition, on more than one occasion, or both. The subjects may often be arranged in different groups, either those naturally occurring such as gender, or those formed by random allocation, for example, when competing therapies are assessed. Such a design is generally referred to as involving repeated measures. Three examples will help in getting a clearer picture of a repeated measures type of study.

  1. Visual acuity data. This example is one already introduced in Chapter 2 , involving response times of subjects using their right and left eyes when light was flashed through lenses of different powers. The data are given in Table 2.3 . Here there are two within subject factors, eye and lens strength.

  2. Field dependence and a reverse Stroop task. Subjects selected randomly from a large group of potential subjects identified as having field-independent or field-dependent cognitive style were required to read two types of words (color and form names) under three cue conditions—normal, congruent, and incongruent. The dependent variable was the time in milliseconds taken to read the stimulus words. The data are shown in Table 5.1 . Here there are two within subjects factors, type and cue, and one between subjects factor, cognitive style.

  Table 5.1   Field Independence and a Reverse Stroop Task

  Note. Response variable is time in milliseconds. N, normal; C, congruent; I, incongruent.

  3. Alcohol dependence and salsolinol excretion. Two groups of subjects, one with severe dependence and one with moderate dependence on alcohol, had their salsolinol excretion levels (in millimoles) recorded on four consecutive days (for those readers without the necessary expertise in chemistry, salsolinol is an alkaloid with a structure similar to heroin). Primary interest centers on whether the groups behaved differently over time. The data are given in Table 5.2

  Sign up to read

  Learn more about book
• 

  eBook - ePub

  Empirical Direction in Design and Analysis

  • Norman H. Anderson(Author)
  • 2001(Publication Date)
  • Psychology Press

    (Publisher)

  Chapter 6 Repeated Measures Design

  Using each subject in a number of experimental conditions has two attractions: one statistical, one substantive. The statistical attraction is that error variability is lower because subjects are their “own controls.” Comparison of treatments A1 and A2 compares two scores from each subject. The main effect of each subject cancels in the difference, thereby freeing the comparison from the main effect of individual differences.

  Using different subjects in different conditions, as in previous chapters, confounds individual differences with conditions. Although randomization resolves this confounding, it does so at the high cost of putting the individual differences in the error term. MSerror typically runs several times smaller in Repeated Measures Designs. Confidence intervals are correspondingly shorter and power is greater. From this statistical standpoint, comparing the same subject across different conditions is most desirable.

  More important is the substantive consideration that psychological process has its locus within the individual. Many investigations focus on the pattern of response across a set of stimuli, but this pattern may be irretrievably confounded with individual differences in response to given stimuli. To study response pattern, therefore, within individual comparison is desirable, perhaps essential.

  Response pattern was the concern in the blame experiment of Figure 1.3 , page 27 of Chapter 1 , which showed the factorial graph of blame as a function of actor’s intent to harm and amount of harm done. In this task, different individuals may show different judgment patterns (e.g., Figure 11.3 , page 318 ). A group graph would hardly be meaningful; repeated measures is essential.

  Analogous concern with response pattern is common in every area of psychology, most notably in perception, but also in diverse other areas from person cognition to operant psychology. Such response patterns are ideally studied with repeated measures.

  Sign up to read

  Learn more about book
• 

  eBook - PDF

  Behavioral Research and Analysis

  An Introduction to Statistics within the Context of Experimental Design, Fourth Edition

  • Max Vercruyssen, Hal W. Hendrick(Authors)
  • 2011(Publication Date)
  • CRC Press

    (Publisher)

  The Repeated Measures Design is useful if you expect large individual differences, for example, to study the effect of a drug on pain tolerance because pain tolerance (DM value) varies widely among individuals. This technique is also used to examine changes in behavior over time, such as the effects of learning on task performance. Where the between-groups design is used to avoid car-ryover effects from one treatment to another, the within-subjects design may be used specifically to investigate those effects. 0 5 6 8 Value of n 11 16 20 0.1 0.2 0.3 Value of r 12 0.4 0.5 0.6 0.7 0.8 FIGURE 4.3 Relationship between n and r 12 . Enter with a value of n and read the your expected value of r 12 that intersects the curve at that point. If your expected value of r 12 exceeds the value obtained from the curve, a matched groups design is preferred. 98 Behavioral Research and Analysis Disadvantages of Repeated Measures Designs One problem with a repeated measures experiment is that the potential for undesirable carryover effects. For example, giving individual subjects a sequence of different drug dosages may produce unforeseeable carryover effects. No matter what order of dosages is used, tolerance or desensiti-zation may occur as a confounding variable because it is not controlled. The randomized groups design is the best for controlling for both known and unknown confounding variables. A second problem is with time-related or order effects; later measures are affected by earlier measures, regardless of the order. These are the effects of fatigue (or practice in tests not designed to study learning). Test scores may improve on a second test, even without feedback on the first test, due to increased familiarity or comfort. Order effects differ from carryover effects in that they are transitory and may be controlled by counterbalancing. Carryover effects may be permanent or semi-permanent, as in drug experiments.

  Sign up to read

  Learn more about book
• 

  eBook - PDF

  Bayesian Methods for Repeated Measures

  • Lyle D. Broemeling(Author)
  • 2015(Publication Date)
  • Chapman and Hall/CRC

    (Publisher)

  1 1 Introduction to the Analysis of Repeated Measures 1.1 Introduction This book presents the Bayesian approach to the analysis of repeated measures. As such, the book is unique in that it is the only one from a Bayesian viewpoint to present the basic ideas about analyzing repeated measures and associated designs. In repeated measures, measurements of the same experimental unit are taken over time or over different study conditions. In a repeated mea-sure study, the main aim is to determine the average value or mean profile of the individual over the range the measurements are observed. Thus, the focus is on the within-individual change of the average response. Repeated measure studies differ from cross-sectional designs, if the same individual is followed over time; on the other hand, with the cross-sectional design, differ-ent individuals appear throughout the observation period. A good example of a cross-sectional study occurs in clinical trials, where one group of sub-jects receives the treatment under study and another a different treatment (or placebo). Of course, it is possible that the same individual can receive the treatment at various time points, followed by receiving another treatment at later time points. Often it is not realistic for the same subject to receive different treatments, but there are scenarios when it is practical. An example of a cross-sectional design is provided by Fitzmaurice, Laird, and Ware (p. 3), 1 who describe a study of a group of girls where the percent body fat is followed up to the time of menarche and another group where the percent body fat is mea-sured at menarche and post-menarche. It is thought that the percent body fat increases at menarche but levels off after a period of approximately four years. A more efficient approach is to follow one group beginning some time before menarche followed by annual fat percent measurements until the expected time of leveling out.

  Sign up to read

  Learn more about book
• 

  eBook - PDF

  Statistics for The Behavioral Sciences

  • Frederick Gravetter, Larry Wallnau(Authors)
  • 2016(Publication Date)
  • Cengage Learning EMEA

    (Publisher)

  In this chapter we extend the ANOVA procedure to single-factor, repeated- measures designs. The defining characteristic of a repeated-measures design is that one group of individuals participates in all of the different treatment conditions. The repeated-measures ANOVA is used to evaluate mean differences in two general research situations: 1. An experimental study in which the researcher manipulates an independent vari-able to create two or more treatment conditions, with the same group of individuals tested in all of the conditions. 2. A nonexperimental study in which the same group of individuals is simply observed at two or more different times. Examples of these two research situations are presented in Table 13.1. Table 13.1(a) shows data from a study in which the researcher changes the type of distraction to create three treatment conditions. One group of participants is then tested in all three conditions. In this study, the factor being examined is the type of distraction. LEARNING OBJECTIVE (a) Data from an experimental study evaluating the effects of different types of distraction on the performance of a visual detection task. Visual Detection Scores Participant No Distraction Visual Distraction Auditory Distraction A 47 22 41 B 57 31 52 C 38 18 40 D 45 32 43 TA B LE 13.1 Two sets of data repre-senting typical examples of single-factor, repeated-measures research designs. (b) Data from a nonexperimental design evaluating the effectiveness of a clinical therapy for treating depression. Depression Scores Participant Before Therapy Immediately After Therapy 6-Month Follow-Up A 71 53 55 B 62 45 44 C 82 56 61 D 77 50 46 E 81 54 55 SECTION 13.1 | Overview of the Repeated-Measures ANOVA 415 Copyright 2017 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).

  Sign up to read

  Learn more about book
• 

  eBook - PDF

  Analysis of Messy Data Volume 1

  Designed Experiments, Second Edition

  • George A. Milliken, Dallas E. Johnson(Authors)
  • 2009(Publication Date)
  • Chapman and Hall/CRC

    (Publisher)

  499 26 Methods for Analyzing Repeated Measures Experiments Like experiments using split-plot designs, experiments utilizing Repeated Measures Designs have structures that involve more than one size of experimental unit. For example, a subject may be measured over time where time is one of the factors in the treatment structure of the experiment. By measuring the subject at several different times, the sub-ject is essentially being “split” into parts (time intervals), and the response is measured on each part. The larger experimental unit is the subject or the collection of time intervals. The smaller unit is the interval of time during which the subject is exposed to a treatment or an interval just between time measurements. Repeated Measures Designs differ from split-plot designs in that the levels of one or more factors cannot be randomly assigned to one or more of the sizes of experimental units in the experiment. In this case, the levels of time cannot be assigned at random to the time intervals, and thus analyzing a repeated measures experiment as though it was a split-plot experiment may not be valid. Because of this nonrandom assignment, the errors corresponding to the respective experimental units may have a covariance matrix that does not conform to the covariance matrix corresponding to experiments for which the usual split-plot analysis is valid. Analyzing a repeated measures experiment as though it was a split-plot experiment is often called a split-plot in time analysis. In Section 26.1, the repeated measures models are described, and the assumptions nec-essary for the split-plot in time analysis of variance to be valid are given. Section 26.2 gives three examples that demonstrate the split-plot in time analysis of variance computa-tions, including the computations of standard errors for making various comparisons between means.

  Sign up to read

  Learn more about book
• 

  eBook - PDF

  Design and Analysis of Experiments with R

  • John Lawson(Author)
  • 2014(Publication Date)
  • Chapman and Hall/CRC

    (Publisher)

  The difference is that the response is measured repeatedly throughout the treatment period instead of once at the end of the treatment period. As an example of a Repeated Measures Design, consider the data in Table 9.8. This is part of the data from a study presented in Diggle et al. (1994) for the purpose of determining how the diet of dairy cows affects the protein in the milk. Seventy-nine Australian cows were randomized to receive one of three diets: barley alone, a mixture of barley and lupins, or lupins alone. The Repeated Measures DesignS 367 protein was measured in a weekly sample of the milk from each cow. Table 9.8 shows the data for the first four weekly samples from the first 10 cows in each group as an illustration of the type of data that results from a Repeated Measures Design. Figure 9.2 shows the trend in average protein over time. It can be seen that the lupins diet results in lowest protein levels, and that the mixed diet, while initially similar to the barley diet in protein, appears to quickly decrease to the level of the lupins diet.

  Sign up to read

  Learn more about book
• 

  eBook - PDF

  Analysis of Variance Designs

  A Conceptual and Computational Approach with SPSS and SAS

  • Glenn Gamst, Lawrence S. Meyers, A. J. Guarino(Authors)
  • 2008(Publication Date)
  • Cambridge University Press

    (Publisher)

  The one-way within-subjects design is also sometimes called a Treatment × Subjects design because all of the levels of the independent variable (the treatment) are crossed with (are administered to) all of the subjects (participants) in the study. 10.3 NATURE OF WITHIN-SUBJECTS VARIABLES Within-subjects variables are those in which participants are measured under each of the conditions (i.e., cases are repeatedly measured). Partic-ipants are thus represented in each and every research condition in the study. Generally, there are two structural forms that this repeated mea-surement can take: (a) it can mark the passage of time, or (b) it can be unrelated to the time of the measurement but simply indicate the condi-tions under which participants were measured. Although these two forms of within-subjects designs do not affect either the fundamental nature of the design or the data analysis, they do imply different research data collection procedures to create the measurement opportunities. 10.3.1 VARIABLES MARKING TIME A within-subjects variable marking the passage of time is one in which the first level of the variable is measured at one point in time, the next level of the variable is measured at a later point in time, the third level of the variable is assessed at a still later period of time, and so on. The most commonly cited example of a time-related within-subjects design is the pretest–posttest study. While not a true experimental design (Campbell & Stanley, 1963; Cook & Campbell, 1979; Shadish, Cook, & Campbell, 2001), partly because there is no control condition, participants are measured at least twice, once at the time designated as the pretest and once at the time designated as the posttest. In such a study the treatment is ordinarily administered between the two measurements and researchers would ordinarily have hypothesized either an increase or a decrease in the characteristic assessed by the dependent variable from the pretest to the posttest.

  Sign up to read

  Learn more about book
• 

  eBook - PDF

  Design and Analysis of Experiments with SAS

  • John Lawson(Author)
  • 2010(Publication Date)
  • Chapman and Hall/CRC

    (Publisher)

  CHAPTER 9 Crossover and Repeated Measures Designs 9.1 Introduction Crossover and Repeated Measures Designs are usually used in situations where runs are blocked by human subjects or large animals. The purpose of crossover designs (COD) is to increase the precision of treatment comparisons by com-paring them within each subject or animal. In a crossover design, each sub-ject or animal will receive all treatments in a different sequence, but the pri-mary aim is to compare the effects of the treatments and not the sequences. Crossover designs are used frequently in pharmaceutical research, sensory eval-uation of food products, animal feeding trials, and psychological research. The primary purpose of Repeated Measures Designs, on the other hand, is to compare trends in the response over time rather than to look at a snapshot at a particular point in time. Each subject or animal receives the same treat-ment throughout the experiment, and repeated measures are taken on each subject over time. Repeated measures experiments are similar to split-plot experiments in that there are two sources of error, treatments are compared to the less precise subject to subject error, and the comparison of trends over time between treatments will be compared to the more precise within subject experimental error. 9.2 Crossover Designs (COD) Crossover designs (CODs) are useful for comparing a limited number of treat-ments, usually from two to six. Since each subject (that will be referred to as the block or whole-plot experimental unit) will receive each treatment se-quentially in time, the number of levels of the treatment factor must remain small, otherwise drop-outs over time will cause problems in the analysis of the data. For this reason CODs are usually not used for factorial treatment plans other than simple 2 2 factorials.

  Sign up to read

  Learn more about book