Assumption of group experimental Design • Variability is intrinsic to human subjects, rather than coming from sources external to the people involved. • The myriad factors —age differences, intelligence, motivation, health status, and so on. • if an experimenter uses a large enough sample and the sample is randomly drawn from the population and randomly assigned to the groups in the research, all those intrinsic factors will be balanced out between experimental and control groups, and the only effects left (aside from experimental errors) will be those of the independent variable or treatment. RANDOMIZED CONTROLLED TRIALS • RCTs are prospective and experimental • (1) the independent (treatment) variable is subject to the controlled manipulation of the investigator and • (2) the dependent (measurement or outcome) variables are collected under controlled conditions. • The independent variable consists of at least two levels, including a treatment group and comparison, or control, group. • RCT refers to random allocation RCT • By definition then, all RCTs are experimental designs. However, not all experimental designs are RCTs, in that there are many non-random ways of placing participants into groups.
• RCTs comparing body-weight–supported
treadmill training to traditional gait training have been published, similar to phase III trials. Cautions About RCTs • For example, if one wanted the strongest possible evidence about whether cigarette smoking causes lung cancer, one would design and implement an RCT comparing a smoking group to a non-smoking group. CON…
• practically impossible to obtain a true random
sample of a population. • Hegde notes that, in random selection, “All members of the defined population are available and willing to participate in the study.” (Not all population in clinic) • All members of a population have an equal chance of being selected for a study. (Ethical consent not obtained) CON…
• In RCTs, patients are (perhaps) randomly
assigned to fixed treatment conditions. However, intervention is individualized. • There is no assumption that results of a group- design experiment are generalizable to any particular individual, • Because results are based on averages and the “average” person probably does not exist. SINGLE-FACTOR EXPERIMENTAL DESIGNS • In 1963, Campbell and Stanley published what was to become a classic work on single-factor experimental design: • Experimental and Quasi-Experimental Designs for Research. • 16 different experimental designs Typology of experimental design
• Campbell and Stanley (1966) developed the typology
of design explored here
• R= Random allocation to group
• O= Observation or measurement • C= Control Group • E= Experimental group • X= exposure of the group to the experimental variable Pretest–Posttest Control-Group Design • The design calls for at least two groups randomly selected (as well as possible) from the population of interest and randomly assigned to one group or another • Performance on the dependent variable(s) is measured and, often, equivalence on the dependent variable(s) is determined statistically. Example • Runeson and Haker used the classic design to study the effect of iontophoresis with cortisone on the treatment of lateral epicondylitis, or tennis elbow. • They compared two groups: a treatment group that received cortisone iontophoresis • and a passive control group that received sham iontophoresis. • They were treated four times over 2 weeks and were measured at the conclusion of the treatment and at 3 and 6 months following the completion of the treatment. • Both groups improved during the study and follow-up period, and there was no statistically significant difference in the response between the two groups, leading the authors to question the use of cortisone iontophoresis in the treatment of tennis elbow. CON… • In clinical research, the pretest–posttest control group design is often altered slightly, as follows: R O X1 O R O X2 O when the researcher does not believe it is ethical to withhold treatment or when two treatments for comparison. A control group receiving a standard treatment can be referred to as an active control group. • Pulvermuller and colleagues used a pretest–posttest control-group design with an active control group in their study of constraint-induced therapy for chronic aphasia after stroke. The treatment group received constraint-induced therapy with intense practice over 10 days. The control group received conventional therapy for aphasia over 4 weeks. Treatment showed significant improvement in a number of communication variables compared with the conventional group. Additional variation • Include taking more than two measurements and using more than two treatment groups. The general notation for this design would be as follows, with the appropriate number of groups and measurement periods: R O O X1 O O ROO X2 O O ROO OO Posttest-Only Control-Group Design
• Researchers use this design when they are not able to
take pretest measurements. • judgments about the effect of the independent variable on the dependent variable. • Like the pretest–posttest control-group design, this design is an RCT and a between-groups design. The Campbell and Stanley10(p25) notation for this design is as follows: RXO RO CON… • Baty et al used this design to compare the effects of a carbohydrate-protein supplement (CHO-PRO) versus a placebo on resistance exercise performance of young, healthy men. • The study was double-blind. The 34 participants were randomly assigned and asked to perform a set of eight resistance exercises. • Performance of the exercises did not vary significantly between groups, although the researchers concluded that the CHO-PRO inhibited muscle damage to a greater degree than did the placebo based on posttreatment measures of myoglobin and creatine kinase. Single-Group Pretest–Posttest Design
• The Campbell and Stanley notation for the
single group pretest–posttest design is as follows: OXO • Unlike the above mentioned designs, the single-group design is neither an RCT nor a between-groups design. EXAMPLE • In a pilot study, Novack et al examined the effects of an in- home occupational therapy program on 20 children with spastic hemiplegic cerebral palsy. • Dependent variables were changes in the Goal Attainment Scaling (GAS), the Pediatric Evaluation of Disability Inventory (PEDI), and the Quality of Upper Extremity Skills Test (QUEST). • They also measured participation amount through a parent self-report. They found that significant changes on the published scales occurred after implementing the home program but found no relationship between participation amount and outcome using the same measures. Nonequivalent Control-Group Design
• Used when a nonrandom control group is available
for comparison. • The Campbell and Stanley notation for this design is as follows: O XO ----------- O O • The dotted line between groups indicates that participants were not randomly assigned to groups. Example • Case-Smith’s study of the effectiveness of a school-based occupational therapy intervention on handwriting. • In this study, occupational therapists in five school districts identified potential participants for the intervention group from among those who had been referred to them for occupational therapy to improve handwriting skills. Then, teachers in those same districts identified potential participants for the comparison group from among students with poor handwriting who had not been referred to occupational therapy. • There were some differences between the two groups that probably resulted from this nonrandom allocation to groups: several children in the occupational therapy group but none in the comparison group were receiving speech language therapy and physical therapy and higher proportions of children in the occupational therapy group were diagnosed with learning or developmental disabilities. Time Series Design
• Used to establish a baseline of measurements
before initiation of treatment to either a group or an individual. • When a comparison group is not available, it becomes important to either establish the stability of a measure before implementation of treatment • The Campbell and Stanley notation for the time series design is as follows: OOOOXOOOO CON… • Ulione used a time series approach to study the impact of a health promotion and injury prevention program at a child development center. Data on the upper respiratory illnesses, diarrhea, and injuries of the children in the development center were collected once a week for 4 weeks before and after the health promotion program. • The four measurements taken before and after the intervention provide a fuller picture of the health status of the children than would single pretest and posttest measures. • Although the time series approach can be applied to group designs, as it was in Ulione’s work, in rehabilitation research it is commonly used with a single-subject approach Repeated Measures or Repeated Treatment Designs • Repeated measures designs are widely used in health science research. • The term repeated measures means that the same participants are measured under all the levels of the independent variable. • In this each participant receives more than one actual treatment. The repeated measures designs are also referred to as with in subjects or within-group designs because the effect of the independent variable is seen within participants in a single group rather than between the groups. When the order in which participants receive the interventions is randomized, this design is considered an RCT with a cross- over design. There are two basic strategies for selecting treatment orders in a repeated measures design with several levels of the independent variable. MULTIPLE-FACTOR EXPERIMENTAL DESIGNS • we wanted to conduct a study to determine the effects of different rehabilitation programs on swallowing function after cerebral vascular accident. • start by selecting 60 patients for study and randomly assigning them to one of the three groups (posture program, oral-motor program, and a feeding cues program). The first independent variable would type of treatment, or group. • Then assume that two different therapists—perhaps a speech-language pathologist and an occupational therapist—are going to provide the treatments. A second independent variable, then, is therapist. • a third question must also be asked in this design: Is one therapist more effective with one type of treatment and the other therapist more effective with another type of treatment? (interaction). 3 × 2 factorial design Completely Randomized Versus Randomized-Block Designs • The Treatment × Therapist • example of a completely randomized design • a randomized-block design is used, in which one of the factors of interest is not manipulable • one of the independent variables is active (manipulable); the other is very often an attribute variable, such as age, or clinical status • Participants would be placed into blocks based on gender and then randomly assigned to treatments. • This arrangement is also known as a mixed design and is extremely common in rehabilitation research. • Most often, the within-subjects factor is the manipulable, active independent variable; all participants receive all the levels of treatment. The within-subjects factor is usually the groups in the research, very often based on an attribute variable (e.g., typically developing versus hearing-impaired children). Mixed design • A mixed, or split-plot, design contains a combination of between-subjects and within-subject factors.