Exam 1 - Study Material (Quiz 2) Flashcards
What is the purpose of research design?
- Provides blueprint/plain - this aids in solving problem/question/hypothesis
- Allows researcher to apply different levels of control
-
We do all of this to say:
- Independent variable really changed the dependent variable and not something else
- To avoid bias
What should you consider when planning “control” in a study?
- Objectivity
- Accuracy
- Feasibility
- CONTROL & FIDELITY
What are some things that affect “control & fidelity” of a deisgn?
- Homogenous sample
- Constancy – Intervention fidelity
- Manipulation
- Randomization
- Validity- Internal
- Validity-External
What is the purpose for control in a study?
- To control for bias
- To ensure Accuracy- Pilot study
- You want to make sure the independent variable really lead to the change in dependent variable and not another variable
What is an Antecedent Variable?
Variables that occur within the subjects prior to the study.
Ex: A patient’s motivation to learn about their medical condition might affect how successful a diabetes education program is for him/her.
What is a ®Intervening, Extraneous or Mediating Variable?
A variable (which the reasearcher doesn’t control) that occurs between the Independent and dependent variable (such as age, gender, diet,etc), and it interferes with the operations of the variables being studied.
Ex: The type of breast cancer treatment patients expereienced.
Describe homogeneous sampling
Sampling that occurs when the researcher’s subjects have similar extraneous variables relevant to the particular study (within the same age group, same gender, etc)
What is generalizability?
Being able to apply research outcomes to other populations when analyzing and discussing the outcomes.
What is Intervention Fidelity in relation to control in a study?
Intervention fidelity involves having consistency in:
- Design - The study is designed to allow an adequate testing of the hypothesis in relation to the underlying theory and clinical processes
- Training - Ongoing training and supervision of the data collecters to assure that the intervention is being delivered as planned and in a similar manner with all subjects
- Delivery - Assessing the intervention is delivered as intended, including that the “dose” is well described for all subjects and that the dose is the same in each group, and that there is a plan for possible problems.
- Receipt - Assurance that the treatment has been received and understood by the subject
- **Enactment - Assessment that the performance of the intervention skills that the subject performs are performed as intended **
What is manipulation in relation to control?
- The addministration of a treatment, program or intervention to only one group within the study and not to the other subjects in the study to test whether a treatment or intervention affects patient outcomes.
- This leads to the creation of the Experimental (the group in which the independent variable is changed) and the control Group (where everything is held constant)
What is randomization in relation to control?
Subjects from the population are obtained in such a manner that each subject has an equal and independent chance of being selected for either group.
This eliminates bias and helps make a representative sample
What are the elements of a research design?
- Participants (WHO)
- Observations (WHAT)
- Selection of subjects (WHERE)
- Role of the investigator
What is internal validity?
- It is the degree to which it can be inferred that the experimental treatment, rather than an uncontrolled condition, resulted in the observed effects
- It asks if it is the independent variable (or something else) that caused or resulted in the chagne in the dependent variable
What are some threats to internal validity?
- History bias
- Selection bias
- Maturation bias
- Testing bias
- Mortality bias
- Instrumentation bias
What is the “history bias” threat to internal validity
It is a specific event outside the study that affects subjects and thus the study results
Ex: An investigator testing the effects of an educational program for young adults on completion of organ donation card designations. During the course of the educational program, an ad featuring a known public figure is released on facebook about he importance of organ donation and becoming an organ donor. The occurence of this information on social media with a public figure engenders a great deal of media and press attention. If the researcher finds an increase in the number of young adults who sign their organ donation cards, the researcher may not be able to conclude that the change in behavior is the result of the teaching program because it may have been influenced by the result of the information on social media.
What is the “selection bias” threat to internal validity?
It is the internal validity threat that arises when pretreatment differences between the experimental group and the control group are present.
Ex: This can occur if the sample was not random or representative of the population being studied, like if a smoking cessation program allowed self-selection of the control or experiemental group
What is the “maturation bias” threat to internal validity?
Developmental, biological or psychological processes that occur within subjects that are independent of the intervention or study variables.
Ex: In a study of new products to stimulate wound healing, one might ask wherther the healing that occured was related to the product or to the natural occurance of wound healing.
What is the “testing bias” threat to internal validity?
It is the effect that occurs from taking the same test repeatedly, influencing a subject’s responses the next time the test is completed.
Ex: Individuals generally score higher when they take a test a second time, regardless of the test.
What does “mortality” refer to in quantitative research?
It is the loss of study subjects from the first data-collection point (pretest) to the second data-collection point (post test)
What is the “mortality bias” threat to internal validity?
This bias can occur if the subjects who remain in the study are not similar to those who dropped out, affecting the results of the study.
Ex: A group of reserachers noted that they had lost subects in all 3 of their groups, and that the lack of statistical significance in their findings may have been related to an inadequate sample size even though the intervention had a positive outcome
What does “instrumentation bias” refer to in quantitative research?
A change in the instruments themselves OR change in the data collection procedures that may account for change in the dependent variable.
Ex: A researcher may wish to study types of thermometers to compare the accuracy of using a tympanic thermometer to other temperature-taking methods. A researcher must check the calibration of the termometers according to the manufacturer’s sepcifications before and after data collection
What is external validity?
- It is the degree to which findings of a study can be generalized to other populations or envrioments
- It questions under what conditions and with what types of subjects can the same results be expected to occur
What are some threats to external validity?
- Selection effects (who)
- Reactive effects (Where-how)
- Measurement effects (How-when-what)
What is “selection” in relation to a quantitative study?
The generalizability of the results to other populations
What is the “selection effects” threat to external validity?
The selection effect refers to how well or not the results of a study can be generalized to other populations.
Ex: If a sample included primarily Caucasion, non-hispanic volunteer couples, inferences about other racial groups should be made with caution
What is “reactivity” in relation to a quantitive study?
The subject’s responses to being studied
What is the “reactive effects” threat to external validity?
- It occurs when subjects respond to the investigator not because of the study procedures, but merely as an independent response to being studied.
- Also known as the Hawthrone effect
What is the “measurement effects” threat to external validity?
The administration of a pretest in a study that affects both the post test results and the generalizability of the findings to other populations
You have to make sure not to “prime” subjects so as not to affect the researcher’s ability to generalize the study to other situations / populations
What are some examples of quantitative design types?
- Experimental
- Quasi-experimental
- Non-experimental
What are some qualitative research types?
- Phenomenology
- Ethnography
- Grounded Theory
- Community Participatory
- Case
What is an experimental group?
The group in an experimental investigation that receives an intervention or treatment.
What is a control group?
The group in an experimental investigation that does not receive an intervention or treatment; the comparison group.
What is a pilot study?
It is a small, simple study conducted as a prelude to a larger-scale study that is often called the “parent study”; it helps to determine the sample size needed for a larger study.
What is “constancy” in refernce to a quantitative study?
It referes to the same methods and procedures (enviromental conditions, timing of data collection, data-collection instruments, etc) are the same for each subject.
What are experimental designs and describe them
- Experimental or Randomized Controlled Trials (RCT), Clinical Trial
- Tests true cause and effect relationships
- Most powerful- threats to internal validity are controlled
- Provide Level II Evidence
What are the Characteristics of True Experimental Study?
- Manipulation
- Control
- Randomization
What is manipulation in reference to an experimental design?
- The provision of some experimental treatment, in one or varying degrees, to some of the subjects in the study
- The process of doing “something” different or a “different”dose to each group
- That something is a treatment, medication or teaching plan
- The effect or difference is measured to assess if the different treatment was better
What is control in reference to an experimental design?
They are measures used to hold uniform or constant conditions under which an investigation occurs
What is the point of “control” in reference to an experimental design?
- Holds conditions constant
- Limits bias that could influence the dependent variable(s)
- Treatments are different
- It is acquired by manipulating the independent variable
What is “randominization” in relation to an experimental design?
- Each subject has equal chance of being in the Experimental or Control group
- It DecreaseS selection bias
- Random assignment is different than random selection
What are the different types of experimental desgins?
- Randomized controled trial
- Solomon Four-group design
- After-only experimental design
How is a randomized controled trial conducted?
- A sample is selected from the population
- Baseline data is collected (pretest)
- Subjects are randomized to either the intervention or contorl group
- Intervention group –> Postintervention data are collected
- Control group –> Postintervetnion data are collected
What do the results from a randomized controled trial indicate?
The trial shows the degree of difference, if any, between the groups at the end of the study, which indicates the confidence the researcher has in a casual link (the intervention that caused the difference) between the independent and dependent variables.
How is a solomon four-group experimental design conducted?
- Sample is selected from the population
- Subjects are randomized
- Basline data is collected (Pre test) –> Intervention group –> postintervention data are collected
- Basline data is collected (Pre test) –> control group –> postintervention data are collected
- No baseline data is collected (No post test) –> Intervention group –> postintervention data are collected
- No baseline data is collected (No post test) –> control group –> postintervention data are collected
How is a nonequivalent control group quasi-experimental design conducted?
- Experimental group –> Baseline data is collected (pretest) –> intervention is applied –> postintervention data are collected
- Control group –> Baseline data are collected (pretest) –> Data are collected again
How is an after-only nonequivalent control group quasi-experimental design conducted?
- Experimetnal group –> intervention is applied –> postintervention data is collected
- Control group –> data is collected
How is a one-group (pretest-postetst) quasi-experimental design conducted?
Experimental group –> baseline data is collected (pretest) –> intervention is applied –> Postintervention data is collected
How is a time series quasi-experimental design conducted?
Experimental group –> data is collected –> data is collected –> intervention is applied –> postintervention data is collected –> postintervention data is collected
What is the “6S” hierachy of pre-appraised evidence?
A model that assits clinicians in their search for the highest level of evidence; the main use for the 6S hierarchy is for efficiently identifying the highest level of evidence to facilitate your search on your clinical question or problem
What are the different levels of organization of evidency in the 6S hierachy?
- Computerized decision support system - Integrates evidence-based clinical infromation into an electronic medical record
- Summaries - Includes clinical practice guidelines, electronic evidence-based textbooks, and evidence-based guidlines that provide recommendations based on high quality evidence
- Synopses of syntheses - Provide a pre-appraised summary of a systematic review. Some include a commentary related to strength of evidence and applicability to a patient population
- Syntheses (Systematic reviews) - These provide a synthesis of research on a clinical topic and use strict methods to search and appraise studies.
- Synopses of studies - A summary of a single study that has been appraised by an expert who provides an overview and appraisal of the study, including a commentary on the study’s context and practice implications.
- Studies - Addresses single stuides that have been pre-appraised.
