N310: Chapter 9 - Quantitative Research

Use of nonexperimental designs in research fulfills what purpose? A) Describing B) Predicting C) Controlling D) Blinding

Ans: A Feedback: A nonexperimental design in research fulfills the purpose of describing. It does not predict, control, or use blinding.

The use of a random numbers table for assigning subjects to groups eliminates which of the following? A) Selection threat B) Intervention fidelity C) Attrition D) Carryover effects

Ans: A Feedback: A selection threat reflects biases stemming from preexisting differences between groups; use of random assignment via a random numbers table eliminates selection threat. Careful researchers pay attention to intervention fidelityóthat is, they take steps to monitor that an intervention is faithfully delivered in accordance with its plan and that the intended treatment was actually received. Longitudinal studies are typically expensive, time-consuming, and subject to the risk of attrition (loss of participants over time), but yield valuable information about time-related phenomena. A crossover design has the advantage of ensuring the highest possible equivalence among the people exposed to different conditions. Such designs are inappropriate for certain research questions, however, because of possible carryover effectsóthat is, when subjects are exposed to two different treatments, they may be influenced in the second condition by their experience in the first. Random assignment does not eliminate intervention fidelity, attrition, or carryover effects.

Which statement infers the advantage to using cross-over designs for quantitative research? A) Enhances equivalence among participants exposed to different interventions. B) Different levels of the intervention are easily compared. C) Delaying the intervention highlights the effects of persuasion. D) Allows all participants to choose the intervention they will receive.

Ans: A Feedback: In crossover designs, people are exposed to more than one experimental condition in random order and serve as their own controls. In cross-over designs, different levels of the interventions are not easily compared, and it does not allow participants of the study to choose interventions they will receive. Delaying the intervention does not highlight the effect of persuasion in a cross-over design.

A study is internally valid to the extent that which of the following has taken place? A) All alternative explanations to the independent variable as the cause of outcomes can be ruled out B) A true experimental design was used C) Intervention fidelity was maintained D) A strong counterfactual was established

Ans: A Feedback: Internal validity is the extent to which it can be inferred that the independent variable is truly causing the outcome. Experiments (or randomized controlled trials, RCTs) involve an intervention (the researcher manipulates the independent variable by introducing an intervention; control (including the use of a control group that is not given the intervention and represents the comparative counterfactual); and randomization or random assignment (with participants allocated to experimental and control groups at random to make the groups comparable at the outset). Intervention fidelity helps to avert biases and gives potential benefits a full opportunity to be realized. A counterfactual is what would have happened to the same people simultaneously exposed and not exposed to a causal factor. The effect is the difference between the two. A good research design for cause-probing questions entails finding a good approximation to the idealized counterfactual.

A review of research reveals two articles that have conflicting results. A national randomized controlled trial found that annual brain scans will allow early detection and treatment of brain cancers, reducing mortality by 65%. An international quasi-experimental trial reported no significant reduction in mortality with annual scans. Based on your review, propose the best recommendation. A) Recommendations from the national study should be followed. B) Recommendations from the international study should be followed. C) There is no rigorous research that can support a recommendation. D) Recommendations should be stratified according to where the research was conducted.

Ans: A Feedback: Methodologically, the main design issue in quantitative studies is whether the research design provides the most valid, unbiased, and interpretable evidence possible. Indeed, there usually is no other aspect of a quantitative study that affects the quality of evidence as much as research design. RCTs are the "gold standard" for intervention studies (Therapy questions) because they yield the highest-quality evidence about the effects of an intervention. Therefore, the best recommendation would be to follow the recommendations from the national randomized controlled trial because it is more authoritative than the quasi-experimental trial.

In an RCT, the most serious threat to internal validity typically is which of the following? A) Mortality B) Selection C) Maturation D) History

Ans: A Feedback: Mortality is the threat that arises from attrition in groups being compared. If different kinds of people remain in the study in one group versus another, then these differences, rather than the independent variable, could account for group differences in outcomes. The selection threat reflects biases stemming from preexisting differences between groups. Selection bias is the most challenging threat to the internal validity of studies not using an experimental design (e.g., nonequivalent control group designs, case-control designs), but can be partially addressed using control mechanisms described in the previous section. The history threat is the occurrence of events concurrent with the independent variable that can affect the outcome. In RCTs, history is not typically a threat because external events are as likely to affect one randomized group as another. The maturation threat arises from processes occurring as a result of time (e.g., growth, fatigue) rather than the independent variable. One-group pretest-posttest designs are especially vulnerable to the maturation threat.

If a researcher wanted to describe the relationship between women's age and frequency of performing breast self-examination, the study would be classified as which of the following? A) Descriptive correlational B) Quasi-experimental C) Longitudinal D) Experimental

Ans: A Feedback: Nonexperimental (observational) research includes descriptive researchóstudies that summarize the status of phenomenaóand correlational studies that examine relationships among variables but involve no intervention or attempt to infer causal connections. Quasi-experiments (trials without randomization) involve an intervention but lack a comparison group or randomization. Longitudinal designs involve data collection at two or more times over an extended period. Experiments (or randomized controlled trials, RCTs) involve an intervention (the researcher manipulates the independent variable by introducing an intervention; control (including the use of a control group that is not given the intervention and represents the comparative counterfactual); and randomization or random assignment (with participants allocated to experimental and control groups at random to make the groups comparable at the outset).

Which of the following situations illustrates control over an external confounding variable? A) Using a script to relay information about the study. B) Randomizing assignment to control treatment groups. C) Allowing for maximum flexibility over where data is collected. D) Choosing a heterogeneous sample of subjects.

Ans: A Feedback: Use of a script to relay information about the study illustrates control over an external confounding variable-study context. Randomizing assignment to control the treatment groups, allowing for maximum flexibility over where data is collected, or choosing a heterogeneous sample of subjects does not illustrate control over an external confounding variable but pertain rather to those intrinsic to study participants.

Constancy of conditions is often enhanced through which of the following? A) Collect data at the same time every day B) Using a crossover design C) Maximizing the external validity of the study D) Avoiding carryover effects

Ans: A Feedback: Various external factors, such as the research environment, can affect outcomes. In carefully controlled quantitative research, steps are taken to minimize situational contaminants (i.e., to achieve constancy of conditions) so that researchers can be confident that outcomes reflect the influence of the independent variable and not the study context. Collecting data at the same time every day is an example of this. A crossover design involves exposing people to more than one treatment. Such designs are inappropriate for certain research questions, however, because of possible carryover effects. External validity concerns inferences about whether relationships found for study participants might hold true for different people, conditions, and settings. These other answers do not enhance constancy of conditions.

Which of the following are distinctive characteristics of a true experimental design? Select all that apply. A) Intervention B) Control C) Randomization D) Correlation

Ans: A, B, C Feedback: A true experimental design is characterized by intervention, control, and randomization. Correlation is not a distinctive characteristic of a true experimental design, correlational design may be used in nonexperimental studies.

A true experiment requires which of the following? Select all that apply. A) Control B) Intervention C) Blinding D) Randomization

Ans: A, B, D Feedback: Experiments (or randomized controlled trials, RCTs) involve an intervention (the researcher manipulates the independent variable by introducing an intervention; control (including the use of a control group that is not given the intervention and represents the comparative counterfactual); and randomization or random assignment (with participants allocated to experimental and control groups at random to make the groups comparable at the outset). Blinding is a means of controlling external sources of bias, but is not required or even feasible for all experiments. By keeping data collectors and others unaware of group allocation or study hypotheses, researchers minimize the risk that other people involved in the study will influence the results.

Cross-sectional designs work best under what research conditions? A) Data collection over an extended period B) Multiple points of data collection C) Describing phenomena at a fixed point D) For follow-up studies

Ans: C Feedback: Cross-sectional designs work best with describing phenomena at a fixed point. Cross-sectional designs involve the collection of data at one time period. The other answers are characteristics of longitudinal studies.

Which of the following are key criteria for making causal inferences about the relationship between two variables? Select all that apply. A) Lack of temporal ambiguity about which variable occurred first B) Statistical confirmation that a relationship between the two exists C) The ability to randomly assign study participants to groups D) The ability to rule out other factors as potential causes of the outcome

Ans: A, B, D Feedback: Key criteria for inferring causality include: (1) a cause (independent variable) must precede an effect (outcome); (2) there must be a detectable relationship between a cause and an effect; and (3) the relationship between the two does not reflect the influence of a third (confounding) variable. The ability to randomly assign study participants to groups is not a key criterion for causality but rather is a method of preventing systematic bias.

The nurse wishes to study the opinions of high school students concerning the availability of health care services at XYZ High School during the past school year. What research design best fits with the study objective? A) Trend study B) Cross-sectional study C) Longitudinal study D) Follow-up study

Ans: B Feedback: A cross-sectional study design would best fit with this study objective. Cross-sectional designs involve the collection of data at one time period, whereas longitudinal designs involve data collection at two or more times over an extended period. Cross-sectional designs are economical and easy to manage, and there is no need for inferring changes over time in this case. In nursing, most longitudinal studies are follow-up studies of clinical populations. Longitudinal studies are best for understanding trends over time.

Which quantitative research design will most strongly support evidence-based practice? A) Factorial Design B) Randomized Control Trial Design C) Correlational Design D) Time-Series Design

Ans: B Feedback: A randomized control trial design will most strongly support evidence-based practice. Randomized controlled trials are the most powerful designs for testing hypotheses of cause-and-effect relationships. RCTs are the ìgold standardî for intervention studies (Therapy questions) because they yield the highest-quality evidence about the effects of an intervention.

Studies that collect data at one point in time are called which of the following? A) Time series B) Cross-sectional studies C) Longitudinal studies D) Crossover studies

Ans: B Feedback: Cross-sectional designs involve the collection of data at one time period, whereas longitudinal designs involve data collection at two or more times over an extended period. In a time-series design, outcome data are collected over a period of time before and after the intervention, usually for a single group. In crossover designs, people are exposed to more than one experimental condition in random order and serve as their own controls.

Using homogeneity as a strategy for controlling confounding variables can reduce which of the following? A) Construct validity B) External validity C) Intervention fidelity D) Internal validity

Ans: B Feedback: External validity concerns inferences about whether relationships found for study participants might hold true for different people, conditions, and settingsóin other words, generalizability. Using a homogeneous sample is easy, but one problem is limited generalizability. Indeed, one problem with this approach is that researchers may exclude people who are extremely ill or incapacitated, which means that the findings cannot be generalized to the very people who perhaps are most in need of interventions. Construct validity involves inferences from the particulars of the study to the higher-order constructs they are intended to represent. If studies contain construct errors, there is a risk that the evidence will be misleading. Intervention fidelity helps to avert biases and gives potential benefits a full opportunity to be realized. Internal validity is the extent to which it can be inferred that the independent variable is truly causing the outcome.

A pretest is to a posttest as which of the following? A) The placebo effect is to the Hawthorne effect B) A baseline measure is to a final outcome measure C) Blinding is to matching D) Attrition is to a mortality threat

Ans: B Feedback: In pretest-posttest designs, data are collected both before the intervention (at baseline) and after it (outcome). The control group can undergo various conditions, including no treatment; an alternative treatment; a placebo or pseudo intervention; standard treatment at different treatment doses; or a wait-list condition. In the Hawthorne effect, various environmental conditions vary to determine their effect on worker productivity. Blinding is another means of controlling external sources of bias. Matching is deliberately making groups comparable on some extraneous variables. Longitudinal studies are typically expensive, time-consuming, and subject to the risk of attrition (loss of participants over time). Mortality is the threat that arises from attrition in groups being compared. If different kinds of people remain in the study in one group versus another, then these differences, rather than the independent variable, could account for group differences in outcomes.

Which of the following is invariably present in quasi-experimental research? A) A control group B) An intervention C) Matching of subjects D) Randomization

Ans: B Feedback: Quasi-experiments (trials without randomization) involve an intervention but lack randomization and, sometimes, a control group. Strong quasi-experimental designs introduce controls to compensate for these missing components. Matching of subjects is a method of controlling confounding variables by consciously forming comparable groups, which is used in some, but not all, quasi-experiments.

Which of the following research designs is weakest in terms of the researcher's ability to establish causality? A) Experimental B) Retrospective case-control C) Prospective cohort D) Quasi-experimental

Ans: B Feedback: Retrospective designs (case-control designs) involve collecting data about an outcome in the present and then looking back in time for possible causes. Experiments are considered the gold standard because they come closer than any other design in meeting the criteria for inferring causal relationships. In prospective cohort designs, researchers begin with a possible cause, and then subsequently collect data about outcomes. Prospective studies are more costly, but much stronger, than retrospective studies. For one thing, any ambiguity about the temporal sequence of phenomena is resolved in prospective research (i.e., smoking is known to precede the lung cancer). Quasi-experiments (trials without randomization) involve an intervention but lack a comparison group or randomization. Strong quasi-experimental designs introduce controls to compensate for these missing components, and thus are generally stronger than non-experimental designs (including retrospective designs).

Which of the following can reduce the statistical conclusion validity of a study? A) Low attrition B) Low power C) Low generalizability D) Low maturation

Ans: B Feedback: Statistical power refers to the capacity to detect true relationships. Statistical power can be achieved in various ways, the most straightforward of which is to use a large enough sample. With small samples, statistical power tends to be low, and the analyses may fail to show that the independent variable and the outcome are relatedóeven when they are, thus reducing statistical conclusion validity. The other answers, although they may affect the study as a whole, do not reduce its statistical conclusion of validity.

The researcher does not have to know in advance which confounding variables have to be controlled for which of the following procedures? A) Matching B) Randomization C) Statistical control D) Homogeneity

Ans: B Feedback: Techniques for controlling subject characteristics include homogeneity (restricting participants to reduce variability on confounding variables); matching (deliberately making groups comparable on some extraneous variables); statistical procedures; and randomization-the most effective method because it controls all possible confounding variables without researchers having to identify them.

In a case-control design, a frequently used method of controlling confounding variables is which of the following? A) Using participants as their own controls B) Matching of cases and controls on confounding variables C) Randomization to groups D) Homogeneity of the sample

Ans: B Feedback: Techniques for controlling subject characteristics include homogeneity (restricting participants to reduce variability on confounding variables); matching (deliberately making groups comparable on some extraneous variables); statistical procedures; and randomizationóthe most effective method because it controls all possible confounding variables without researchers having to identify them.

The nurse is designing a research study to assess the effectiveness of two wound healing dressings. What key design feature will be most important to address in this type of study? A) Intervention B) Comparisons C) Blinding D) Location

Ans: B Feedback: The key design feature would be to address the use of comparisons. An intervention is a treatment. Location of the study would not be most important with regards to the design feature. Blinding refers to keeping certain information from research participants to minimize affecting the results via expectation bias.

In a nonequivalent control group design, the most serious threat to internal validity is which of the following? A) Testing B) Selection C) Maturation D) History

Ans: B Feedback: The selection threat reflects biases stemming from preexisting differences between groups. Selection bias is the most challenging threat to the internal validity of studies not using an experimental design (e.g., nonequivalent control group designs, case-control designs), but can be partially addressed using control mechanisms described in the previous section. The history threat is the occurrence of events concurrent with the independent variable that can affect the outcome. The maturation threat arises from processes occurring as a result of time (e.g., growth, fatigue) rather than the independent variable. RCTs have a clear superiority for testing causal hypotheses. Internal validity is the extent to which it can be inferred that the independent variable is truly causing the outcome.

Which of the following statements best illustrates causality? A) Individuals who have suffered a myocardial infarction (MI) are at higher risk for a second MI within the next 5 years. B) Prolonged sun exposure is associated with higher rates of skin cancer. C) Hiatal hernias are diagnosed more frequently in elderly individuals. D) Taking time away from work leads to fewer stress-related illnesses.

Ans: B Feedback: The statement, "Prolonged sun exposure is associated with higher rates of skin cancer" best illustrates causality. The other statements involve a greater likelihood that multiple variables could explain the effect, whereas the link between prolonged sun exposure and skin cancer is much stronger.

What is the strongest method of controlling for intrinsic (subject) factors? A) Statistical control B) Randomization C) Matching D) Homogeneity

Ans: B Feedback: The strongest method of controlling for intrinsic (subject) factors is randomization. It is the most effective method of controlling participants' characteristics. Homogeneity occurs when people who are similar in respect to confounding variables are included in the study. Matching involves consciously forming comparable groups. Statistical control is the ability to have control over confounding variables statistically.

The research design for a quantitative study involves decisions with regard to which of the following? Select all that apply. A) Which conceptual framework to use B) Whether there will be an intervention C) What types of comparisons will be made D) How many times data will be collected

Ans: B, C, D Feedback: The research design is the overall plan for answering research questions. In quantitative studies, the design designates whether there is an intervention, the nature of any comparisons, methods for controlling confounding variables, whether there will be blinding, and the timing and location of data collection. Choosing a conceptual framework is not a significant part of the research design for quantitative studies but is more important in qualitative studies.

Using your knowledge of threats to internal validity, which research design will be most susceptible? A) Pretest-Posttest B) Cross-over C) Correlational D) Factorial

Ans: C Feedback: A correctional research design is most susceptible to threats to internal validity. With quasi-experiments and correlational studies, there are competing explanations, which are sometimes called threats to internal validity.

One weakness associated with cause-probing correlational research is which of the following? A) Artificiality of the settings in which it occurs B) Difficulty in linking the research to a theoretical framework C) Problem of self-selection into groups D) Inability to generalize the findings beyond the sample

Ans: C Feedback: Correlational studies are susceptible to faulty interpretation because groups being compared have formed through self-selection (also called selection bias). When researchers study the effect of a cause they cannot manipulate, they use correlational designs that examine relationships between variables. A correlation is an interrelationship or association between two variables, that is, a tendency for variation in one variable to be related to variation in another. Correlations can be detected through statistical analyses.

Applying your knowledge of random assignment, which statement is correct? A) Random assignment is accomplished with random sampling. B) Grouping participants with similar features together is the best way to achieve random assignment. C) Random assignment ensures that the study is a true experiment. D) Recruiting participants from significantly different neighborhoods results in random assignment.

Ans: C Feedback: Experimental designs involve placing participants in groups at random. Through randomization (or random assignment), every participant has an equal chance of being included in any group. Randomization is a critical characteristic of a true experimental study. The other answers are not true of random assignment. Random sampling is a method of selecting people for a study and is not used to accomplish random assignment.

When participants' behaviors are affected not by the treatment per se but by their knowledge of participating in a study, interpretation of the findings is complicated by the influence of which of the following? A) Treatment effect B) History threat C) Hawthorne effect D) Selection threat

Ans: C Feedback: Hawthorne effect, in which various environmental conditions vary to determine their effect on worker productivity. The history threat is the occurrence of events concurrent with the independent variable that can affect the outcome. The selection threat reflects biases stemming from preexisting differences between groups. Temporal ambiguity, not treatment effect, the cause must precede the effect.

A study that followed, over a 20-year period, 500 users and 500 non-users of oral contraceptives to determine if there were any long-term side effects would be which of the following? A) Time series B) Retrospective study C) Prospective study D) Crossover study

Ans: C Feedback: In prospective cohort designs, researchers begin with a possible cause, and then subsequently collect data about outcomes. Retrospective designs (case-control designs) involve collecting data about an outcome in the present and then looking back in time for possible causes. In a time-series design, outcome data are collected over a period of time before and after the intervention, usually for a single group. In crossover designs, people are exposed to more than one experimental condition in random order and serve as their own controls.

A one-group pretest-posttest design is an example of which of the following? A) A crossover design B) A true experimental design C) A quasi-experimental design D) A retrospective design

Ans: C Feedback: Quasi-experiments involve an intervention but lack randomization and sometimes even a comparison group. Strong quasi-experimental designs introduce controls to compensate for these missing components. In crossover designs, people are exposed to more than one experimental condition in random order and serve as their own controls. Experiments involve an intervention; control; and randomization or random assignment. Retrospective designs involve collecting data about an outcome in the present and then looking back in time for possible causes.

Quasi-experimental research designs lack what feature found in true experimental research? A) Control groups B) Pretests C) Randomization D) Placebos

Ans: C Feedback: Randomization is not seen in a quasi-experimental research design. Quasi-experimental studies may have control groups, pretests, and placebos.

A research proposal states that the objective is to, "explore the incidence of homeopathic health practices on Iowa Native Reserves in Oklahoma." In describing the design to be used for conducting the study, the researchers plan to use a nonexperimental design. What can you conclude from this proposal? A) The choice of a nonexperimental design is appropriate for the research objective. B) Researchers should consider using a quasi-experimental design to meet their objective. C) This research may not be answered completely if a quantitative design is used. D) The population to be studied needs to be expanded to include all Native Americans in Oklahoma to avoid a small sample size.

Ans: C Feedback: The research may not be answered completely if a quantitative design is used, because the research question is not amenable to quantitative research. There are various reasons for doing a nonexperimental study, including situations in which the independent variable inherently cannot be manipulated (Prognosis questions) or in which it would be unethical to manipulate the independent variable (some Etiology questions). Experimental designs are also not appropriate for Descriptive questions.

When addressing control of confounding variables, which of the following best describes the question to be answered by the researcher? A) Who needs to know sensitive information about the research details? B) When will the data on the variables be collected? C) What other variables may influence the results? D) Where will data collection occur?

Ans: C Feedback: When addressing control of confounding variables, the researcher needs to answer the question, "What other variables may influence the results?" The other questions do not influence the control of confounding variables.

What study design flaw may lead to a failure to achieve statistical significance? A) Variables precisely defined B) Adequate exposure to the intervention C) Small sample size D) Cross-sectional data collection

Ans: C Feedback: With small samples, statistical power tends to be low, and the analyses may fail to show that the independent variable and the outcome are relatedóeven when they are. Variables that are precisely defined and have adequate exposure to the intervention and cross-sectional data collection lead to statistical significance.

Which of the statements below best illustrates the temporal criterion needed for a causal relationship? A) Skin cancer occurs because of genetic predisposition. B) Kidney disease develops as a result of pesticide exposure. C) Hepatitis C occurs in populations with substance abuse histories. D) Following vaccination for varicella, rates of varicella infection are lower.

Ans: D Feedback: A cause must precede an effect in time. The answer "Following vaccination for varicella, rates of varicella infection are lower" best establishes that the suspected cause (vaccination) comes before the observed effect (decrease in varicella infection rates)..

The nurse plans a study comparing the occurrence of anxiety disorders in military personnel deployed overseas with those who served strictly within the borders of the United States. What research design should be selected for this study? A) Experimental B) Quasi-Experimental C) Cohort D) Nonexperimental

Ans: D Feedback: A nonexperimental design should be selected for this study. When researchers do not intervene by controlling the independent variable, the study is nonexperimental, or, in the medical literature, observational.

Which of the graphic representations illustrates a randomized experimental interventional design with pre- and posttest? A) R X O X B) R O R X C) R X O O D) R O X O

Ans: D Feedback: Experimental designs can be depicted graphically using symbols to represent features of the design. In these diagrams, the convention is that R stands for randomization to treatment groups, X represents receipt of the intervention, and O is the measurement of outcomes. So, for example, a pretest-posttest design would be depicted as follows: R O X O. The other answers do not illustrate the correct order of elements.

The use of a diverse sample of study participants in multiple sites might affect which of the following? A) Replicability of the study B) The ability to use randomization C) The ability to use blinding as a strategy D) The study's external validity

Ans: D Feedback: External validity concerns inferences about generalizability-whether findings hold true over variations in people, conditions, and settings. Having a diverse sample of study participants in multiple sites would likely enhance the study's generalizability, and thus its external validity. The other answers would not necessarily be affected by a diverse sample of study participants in multiple sites.

An important function of a rigorous research design in a quantitative study is to have control over which of the following? A) Outcome variables B) Mediating variables C) Carryover variables D) Confounding variables

Ans: D Feedback: Key criteria for inferring causality include: (1) a cause (independent variable) must precede an effect (outcome); (2) there must be a detectable relationship between a cause and an effect (mediating); and (3) the relationship between the two does not reflect the influence of a third (confounding) variable. A crossover design has the advantage of ensuring the highest possible equivalence among the people exposed to different conditions. Such designs are inappropriate for certain research questions, however, because of possible carryover effects. Outcome variables are not to be controlled but are the effect that is under study. Mediating variables are those through which another variable has an effect on the outcome variable. They do not need to be controlled.

Which situation best reflects "attrition" in quantitative research? A) Control and intervention groups are very different in age ranges. B) Subjects receive promotional materials about the benefits of the intervention. C) Adolescent subjects improve in body hygiene practices over 4 years. D) Control group subjects drop out of the study when their disease process does not improve.

Ans: D Feedback: Mortality is the threat that arises from attrition in groups being compared. The most severely ill patients might drop out of the experimental condition because it is too demanding.

Which of the following is the most effective method for controlling participant factors? A) Using a homogeneous sample B) Statistical control C) Matching subjects D) Randomization

Ans: D Feedback: Techniques for controlling subject characteristics include homogeneity (restricting participants to reduce variability on confounding variables); matching (deliberately making groups comparable on some extraneous variables); statistical procedures; and randomizationóthe most effective method because it controls all possible confounding variables without researchers having to identify them.

To study the effects of Healing Touch; what is the best choice to minimize confounding variables? A) Offer a small payment for participants' time. B) Offer all participants a Healing Touch session when data collection is completed. C) Offer to enroll only individuals who do not have any medical conditions. D) Offer another activity that gives similar time and attention to control group participants.

Ans: D Feedback: The best way to minimize confounding variables is to offer another activity that gives similar time and attention to control group participantsóknown as a placebo. Otherwise, the participant's knowledge of not receiving a treatment may affect the study's results. None of the other answers would control for this potential confounding variable.

The threat to internal validity that occurs when external co-occurring events or conditions affect outcomes is the threat known as which of the following? A) Maturation B) Selection C) Testing D) History

Ans: D Feedback: The history threat is the occurrence of events concurrent with the independent variable that can affect the outcome. The maturation threat arises from processes occurring as a result of time (e.g., growth, fatigue) rather than the independent variable. The selection threat reflects biases stemming from preexisting differences between groups. RCTs have a clear superiority for testing causal hypotheses. Internal validity is the extent to which it can be inferred that the independent variable is truly causing the outcome.