Nurs 413 Ch.6 & 7

Quasi experimental designs include which on the following essential components: (select all that apply) 1. randomization, control group, and manipulation of IV 2. randomization, and control group 3. comparison group and manipulation of IV 4. randomization and manipulation of IV

3. comparison group and manipulation of IV 4. randomization and manipulation of IV

Nonexperimental designs are used for the purpose of:

1. describing a phenomenon in detail 2. explaining relationships and differences among variables 3. predicting the relationships and differences among variables

3 quasi-experimental designs

1. nonequivalent control group pretest-posttest 2. time series designs 3. preexperimental designs

Two aims of predictive designs

1. researchers attempt to determine the amount of variance in an outcome variable that can be explained by multiple predictive variables 2. determining group membership

RCTs are characterized by the following

1. they involve a large # of subjects, often from diverse geographical areas 2. there are strict guidelines for including subject in a study 3. subjects are randomly assigned to interventions and control groups 4. groups must be comparable on key characteristics at baseline 5. the intervention is consistently implemented to all subjects in the experimental group following a very rigidly defined protocol for implementation 6. all subjects in both groups are measured on the DV using the same method of measurement at the same points in time

6 types of true experimental designs

1. two-group pretest-posttest 2. two group, posttest only 3. Solomon four group 4. multiple experimental groups 5. factorial 6. crossover designs

Any study that involves collecting data at multiple points in time is a longitudinal study. True or False

False

In a factorial design, only one group of subjects is required because they act as their own control. True or false

False

In retrospective designs, also known as ex post facto designs, the researchers manipulates the IV. True or false

False

The threat of mortality is greater in cross-sectional designs than in longitudinal designs. True of False

False

Probability

Likelihood or chance that an event will occur in a situation

Can causal statement be made with correlational designs?

No because they do not involve manipulation of IV

Why is the Soloman four group a superior design than the two group pretest posttest?

Reduction of testing threat; also superior to two group posttest only because it minimizes selection bias

What agency is committed to the support of translational research?

The Agency for Healthcare Research and Quality (AHRQ)

All experiments must include a minimum of two groups of subjects. True or False

True

Cohort comparison studies can save time because more than one group of subjects is studied. True or False

True

Descriptive data are usually cross-sectional and can be controlled through surveys and questionnaires

True

Patients who are not in the intervention group must receive the standard of care. True or False

True

The Solomon Four group design is more effective at controlling for the threat of testing than is the two group pretest posttest design. True or False

True

The purposes of nonexperimental designs are to describe, explain, and predict relationships.

True

What is considered to be the classic experimental design?

Two group pretest-posttest

Elements to consider when appraising quantitative designs

What type of quantitative design is being used? Can causality be inferred? Did researchers use the highest level of design possible to answer the research questions? What strategies did the researcher use to control for the effects of extraneous variables? What threats to internal and external validity might be present, and were any strategies used to reduce them? Were there any ethical concerns about the design?

Survey design

a descriptive design involving data obtained through subjects' self report

Disadvantage of multiple group design

a large # of subjects are needed to detect differences across multiple groups

Follow-up

a longitudinal design used to follow subjects into the future

Preexperiemental

a posttest only design that involves manipulation of the IV but lacks control for extraneous variables

One group posttest only design

a preexperimental design involving one group and a posttest with little control for extraneous variables

Nonequivalent groups posttest only design

a preexperimental design involving two groups measured after an intervention with little control for extraneous variables

Nonequivalent control group pretest posttest design

a quasi experimental design where two groups are measured before and after an intervention

Time series design

a quasi-experimental design where one group is measured prior to administering the intervention and then multiple times after the intervention

nonexperimental design

a research design that lacks manipulation of the IV and random assignment

Interaction of treatment and setting

a threat to external validity when an intervention conducted in one setting cannot be generalized using the same intervention

Interaction of treatment and history

a threat to external validity when historical events affect the intervention

Construct validity

a threat to external validity when the instrument does not accurately measure the theoretical concepts

Interaction of treatment with selection of subjects

a threat to external validity where the IV might not affect individuals the same way

Testing

a threat to internal validity when a pretest influences the way subjects respond on a posttest

Maturation

a threat to internal validity when subjects change by growing or maturing

Selection bias

a threat to internal validity when the change in the DV is a result of to characteristics of the subjects before they entered the study

History

a threat to internal validity when the dependent variable is influenced by an event that occurred during the study

Mortality

a threat to internal validity when there is a loss of subjects before the study is completed

Trend

a type of longitudinal design to gather data different samples across time

The researcher's ability to manipulate or regulate extraneous variables is known as: a. control b. manipulation c. bias d. probability

a. control

The degree to which the results of studies can be generalized to other individuals, settings, or time is: a. external validity b. construct validity c. internal validity d. statistical conclusion validity

a. external validity

When designing a study, which of the following should the researcher consider: (select all that appyl) a. research question b. review of the literature c. theoretical framework d. study purpose

a. research question b. review of the literature c. theoretical framework d. study purpose

Study validity

ability to accept results as logical, reasonable, and justifiable based on the evidence presented.

Control

ability to manipulate, regulate, or statistically adjust for factors that can affect the dependent variable

Community-based participatory action research (CBPAR)

active involvement of community members throughout the research process

In experimental designs, researchers

actively manipulate the independent variable (causal variable) to determine its effect on the dependent variable

Advantages of multiple group design

allows researchers to compare the effect of different interventions on the DV

True causality can only be established with

an experimental design

Two group posttest only design

an experimental design when subjects are randomly assigned to an experimental or control group and measured after the intervention

Solomon four group design

an experimental design with four groups, some receive the intervention, other serve as control; some are measured before and after, others only measured after the intervention

An appropriate research design will reflect

an integration of the theoretical and empirical literature that was presented in the review of the literature section of the study

Threats

are forces that can change the results of the study

In nonexperimental designs, researchers

are observers noting the occurrence of the variables of interest and trying to determine relationships and differences.

A researcher plans to observe children in a kindergarten class. Students have always been told to be on their best behavior when guests are present in the classroom. What is the greatest threat to external validity: a. construct validity b. Hawthorne effect c. selection d. interaction of treatment setting

b. Hawthorne effect

Experimental designs have control groups. Quasi experimental designs have which of the following: a. control groups b. comparison groups c. extraneous groups d. peer groups

b. comparison groups

When a researcher assign subjects to groups by tossing a coin, the researcher is using: a. random selection b. random assignment c. bias d. within-groups design

b. random assignment

Rank the evidence generated from the following designs from lowest to highest: a. experimental designs b. nonequivalent control group pretest posttest c. one group posttest only d. nonequivalent groups posttest only

c, d, b, a

During a study examining nurses' job satisfaction, the union decides to hold a strike. This is what type of threat to internal validity: a. selection bias b. mortality c. history d. testing

c. history

Advantage of longitudinal studies

can be used to test cause and effect

One problem with crossover studies

carryover effect, where even though the subjects are asked to stop performing the 1st intervention, the may continue to use it during the 2nd intervention

Within-groups design

comparisons are made about the same subjects at two or more points in time or on two or more measures

To control for the threat of instrumentation, researchers need to ensure

comprehensive training of all data collectors

Control group is one strategy to

control for extraneous variables

What can help limit maturation threat to internal validity?

control group

Model-testing

correlational design to test a hypothesized theoretical model

Disadvantage of longitudinal studies

costs in following subjects over an extended period of time

Which of the following is not a purpose of nonexperimental designs: a. describe phenomenon b. explain relationships c. predict relationships d. examine causality

d. examine causality

Two general ways the nonexperimental designs can be categorized

descriptive and correlational

Comparative design

descriptive design that compares 2 or more groups or variables

3 types of correlational designs

descriptive-correlational predictive correlational model-testing

experimental design

design involving random assignment to group and manipulation of the IV

Longitudinal designs

designs used to gather data about subjects at more than one point in time

Limitation of cross-sectional design

difficult for researchers to make claims about cause and effect

Attrition rate

dropout rate

Advantages of cross-sectional designs

easier to manage and are more economical; data is collected only one time so the threats of mortality, maturation, and testing are minimized

Critical principles for CBPAR include

engaging in collaborative, equitable partnerships, building on resources and using dynamic processes where ideas flow between researchers and community members

Factorial design

experimental design allowing researchers to manipulate more than one intervention

Multiple experimental groups design

experimental design using two or more experimental groups with one control group

Crossover design

experimental design using two or more treatments and subjects receive both treatments in a random order

Crossover

experimental design using two or more treatments; subjects receive treatments in random order

Quantitative designs used to examine relationship among variables are categorized as:

experimental or nonexperimental

Extraneous variable

factors that interfere with the relationship between the IV and DV

In nonexperimental designs, researchers manipulate the IV to determine cause and effect. True or False

false

Researchers use correlations to determine if there are differences between two groups

false

Repeated testing can cause _________ with the test itself and ______ may reflect subjects' abilities to remember how the questions were ________ previously rather than reflecting __________ and _________.

familiarity answers answered current knowledge beliefs

The threat of history can be decreased by

having a control group that is exposed to the "event" but does not receive the intervention

Evaluating the quality of quantitative evidence requires the nurse to

identify potential threats to validity, evaluate the seriousness of those threats, and determine if results are valid for application to patient care

Retrospective designs are often used

in epidemiological studies

Panel design

longitudinal design where the same subjects provide data from different samples across time

Researchers must be able to ________ the IV for a design to be considered experimental

manipulate

Cohort comparison

more than one group is studied at the same time so that conclusion about a variable over time can be drawn without spending as much time

When outcomes have many causes, its called

multicausality

Correlational design

nonexperimental design used to study relationship between two or more variables

Exploratory design

nonexperimental design used when little is known about a phenomenon

Cross-sectional

nonexperimental designs used to gather data from a group of subjects at only one point in time

Descriptive design

provide a picture of a situation as it is naturally happening without manipulation of any of the variables

If the attrition rate is high the author of the article should

provide an analysis and explanation for the dropout rate

To be considered a true experimental design, 3 features must be included:

randomization, control, and manipulation

Randomization happens in 2 ways

randomly select subject from the target population randomly assign subjects to groups

Potential disadvantage of the two group pretest-posttest design

repeated testing threats

Retrospective design

research design when researchers look back in time to determine possible causative factors

Quasi-experimental designs

research designs involving the manipulation of the IV but lacking either two groups or a control group

Translational research

research for the purpose of linking research findings to the point of care

Health services research

research involving phenomena related to the delivery of health care

ex post factor

retrospective research design

Case control

retrospective study where research begins with people who already had the disease

Common threats to internal validity include:

selection bias history maturation testing instrumentation mortality statistical conclusion validity

The term longitudinal is reserved for

studies in which data are gathered over extended period of time rather than in just a few hours or days

Prospective designs

studies over time with presumed causes that follow subjects to determine if the hypothesized effects actually occur

Double-bind experimental designs

studies where subjects and researchers are unaware if they're experimental interventions or standard of care

Between-group design

study design where two groups of subjects can be compared

Two-group pretest-posttest designs

subjects are randomly assigned to the experimental or control group, and measured before and after the intervention

Ransom assignment

subjects have an equal chance of being in either the treatment or the control group

Hawthorne effect

subjects' behaviors affected by personal values or desires to please the experimenter

Disadvantage of the two group posttest only design

susceptible to the threats of selection bias

Random sampling

technique for selecting elements whereby each has the same chance of being selected

The difference between pre and quasi experimental designs is

that quasi, while lacking some elements of an experiment, use other strategies to control for extraneous variable

Because retrospective designs are not experimental, a disadvantage is

that researchers cannot say definitively that to IV caused the DV; instead they can conclude that there is an increased likelihood or probability that the IV caused the DV

Manipulation

the ability of researchers to control the independent variable

Statistical conclusion validity

the degree that the results of the statistical analysis reflect the true relationship between the IV and DV

Internal validity

the degree to which one can conclude that the IV produced changes in the DV

External validity

the degree to which the results of the study can be generalized

Reactivity

the influence of participating in a study on the responses of subjects

Experimental designs are used for

the purpose of examining causality

Nonexperimental designs can be used for

the purposes of describing a phenomenon in detail, explaining relationships and differences among variables, and predicting relationships and differences among variables

Causality

the relationship between a cause and its effect

The major difference between nonexperimental and experimental designs is

the role of the researcher

Nonexperimental studies differ from panel studies in that

the samples are not drawn from the general population, but instead samples are selected because they have a specific characteristic or condition that researchers are interested in studying

Randomization

the selection, assignment, or arrangement of elements by chance

In health-related experimental designs, the control group of subject usually receives:

the standard of care, but does not get the intervention

RCT (randomized controlled trial)

the study is clinical in nature rather than a specific type of design

Who is the control group in a crossover design?

the subjects serve as their own control group

Selection bias con be minimized somewhat by

the use of random assignment to groups

Nonexperimental designs are important when

there is little information known about a particular phenomenon, when it would be unethical to manipulate the IV, or when it is not practical to conduct an experiment.

Effects of selection

threat to external validity when the samples does not represent the population

Instrumentation

threat to internal validity when there are inconsistencies in data collection

Nonexperimental design can be used to develop and test theories

true

Descriptive correlational design

used to explain the relationship among the variables or groups using a nondirectional hypothesis

Researches can control for low reliability of the measure by

using well-established and well-designed instruments

Covary

when change in one variable is associated with change in another variable

Bias

when extraneous variables influence the relationship between the independent and dependent variables

Predictive correlational design

when researchers hypothesize which variables are predictors or outcomes

Type II error

when researchers inaccurately conclude that there is no relationship between the IV and DV