Research Test 2

Describing the sample should include:

-Demographic characteristics of the sample -Inclusion/exclusion criteria -Sampling strategy used -Screening, recruitment, and enrollment protocol

Disadvantages of Stratified Random Sampling

-Difficult to obtain a population list containing complete critical variable information. -Time consuming -Enrolling proportional strata is challenging -A large-scale study is costly and takes time

Inferential statistics

-Allow the testing of hypotheses using data obtained from probability and nonprobability samples -A parameter is a characteristic of a population -A statistic is characteristic of a sample

Quasi-experimental designs

-Also test cause-and-effect relationships -Subjects not randomly assigned to groups -Randomization may not be possible, or there may not be a control group*** -Lack of control makes evidence provided by quasi-experimental design less convincing -Provide Level III evidence

Accessible population

-Available population that meets the criteria for the target population Ex: All cholecystectomies performed on female patients admitted to hospital X in Southern CA

What is another name for retrospective studies? Repeated measures studies Case-control studies Cohort studies Ex post facto studies

-Case-control studies -Ex post facto studies -Causal-comparative -Comparative studies

Data collection methods utilizes operational definitions to:

-Clearly define variables/concepts of interest -Translate variables/concepts into measurable values for data collection -Answer the question "How will the researcher measure variables?'

Relationship/Difference studies (non-experimental studies)

-Correlational studies -Developmental studies: cross-sectional, longitudional or prospective, and retrospective or ex post facto

Where is validity and reliability reported?

-Methods section

Homogenous sample

-Participants are similar in regard to extraneous variables (that are relevant to the particular study) -Limits generalizability or the potential to apply the results of a study to other populations.

Advantages of Stratified Random Sampling

-Representativeness of the sample is enhanced -It is able to make comparisons among subsets -A disproportionately small stratum is able to be oversampled to adjust for underrepresentation.

Power Analysis

-The mathematical procedure to determine the # for each group of the study. -Used to determine the right sample size

Intervention

-The observer provokes actions from those who are being observed

Open-ended questions are used when:

-The researcher does not know all of the possible alternative responses.

Purpose of Sampling

-To increase the efficiency of a research study

Weakness of Quasi-experimental designs

-Unable to demonstrate clear cause-and-effect relationships

Advantages of Retrospective-Ex Post Facto-Case control studies

-Useful in testing theoretical models of how variables work together in a group in a particular situation -Important designs when randomization, control, and manipulation are not appropriate or impossible.

Critical appraisal of this section considers threats to:

-Validity -Source of bias

Population

-Well-defined set with specific properties -May be humans, medical records, specimens

Close-ended questions are used when:

-When the question has a finite number of responses and the respondent is to choose the one closest to the correct response

Inclusion criteria is also called

-eligibility criteria

A study has external validity if:

-the results of the study can be generalized to other populations and settings

A Z score of +1 indicates a score that is: Too low to be caused by chance alone 1 standard deviation (SD) above the mean Skewed to the right 1 SD below the mean

1 standard deviation (SD) above the mean Rationale (CORRECT): 2. A Z score of +1 means that the observation is 1 SD above the mean.

levels of measurement for descriptive & inferential statistics

1. Nominal 2. Ordinal

A researcher systematically chooses a sample of 10 patients from a list of 600. The sampling interval would be: 30 6 60 15

60 Rationale (CORRECT): 3. The researcher would choose every 60th patient on the list.

Unstructured observational methods

Have some structure Usually involve collecting descriptive information Participant observation uses field notes to record the activities, as well as the observer's interpretations Anecdotes usually focus on the behaviors of interest and illustrate a particular point

Type II error

Sample size is too small, so differences may not be detected

Which type of study is a comparative study? Survey Correlational Developmental Experimental

Survey Rationale (CORRECT): 1. A comparative study is a survey.

Validity is concerned with which type of error?

Systematic error

Objectivity is derived from

review of literature and development of theoretical framework

Which of the following are most likely operational definitions of variables? (Select all that apply.) Verbalized satisfaction with care received Quality of life Pain rated on a scale of 1 to 10 Heart function

Verbalized satisfaction with care received Pain rated on a scale of 1 to 10

The literature review should be presented so that the reader can...

judge the objectivity of the research question and therefore judge whether the design chosen matches the research question.

All data collection methods rely on the ability of the researcher to

standardize procedures to increase data accuracy and reduce measurement error

Objectivity in conceptualization of the research question is accomplished through...

the theoretical framework and thorough review of literature

Physiological measurement involves

-the use of technical instruments to collect data about patients' physical, chemical, microbiological, or anatomical status.

Which reliability measure is considered to be a test of equivalence? Split-half reliability Test-retest reliability Parallel, or alternate-form, reliability Cronbach's alpha

arallel, or alternate-form, reliability Rationale (CORRECT): 3. Parallel, or alternate-form, reliability is considered a test of equivalence.

Which critiquing criterion would apply to data-collection methods in general? Were the observers trained to minimize any bias? Are operational definitions provided and clear? Is there clear indication that the subjects understood the task and the questions? Is there evidence that subjects were able to answer the questions?

Are operational definitions provided and clear? Rationale (CORRECT): 2. This critiquing criterion would be applicable to data-collection methods in general.

Which level of measurement uses the mode? (Select all that apply.) Nominal Ordinal Interval Ratio

Nominal Ordinal Interval Ratio

Which level of measurement focuses on classification? Nominal Ordinal Interval Ratio

Nominal Rationale (CORRECT): 1. Nominal measurement focuses on classification.

What are levels of measurement in ascending order? Interval, ordinal, nominal, ratio. Ordinal, nominal, interval, ratio. Nominal, ordinal, interval, ratio. Nominal, ordinal, ratio, interval.

Nominal, ordinal, interval, ratio.

Which measure of variability is the most unstable? Mode Standard deviation Percentile Range

Range Rationale (CORRECT): 4. The range is most affected by high and low scores.

Which measure of variability indicates the range of the middle 50% of the scores? Semiquartile range SD Percentile Parameter

Semiquartile range Rationale (CORRECT): 1. The semiquartile range indicates the range of the middle 50% of the scores. It lies between the upper and lower quartiles, the upper quartile being the point below which 75% of the scores fall and the lower quartile being the point below which 24% of the scores fall.

Content validity

The concern is whether the measurement instrument and the items it contains are representative of the content domain that the researcher intends to measure. *Face validity is a subtype of content validity

What is external validity? The use of expert opinion to judge the accuracy of an instrument The degree to which findings of a study can be generalized to other populations The degree to which it can be inferred that the experimental treatment resulted in the observed effects The degree to which an instrument is said to measure a theoretical construct or trait

The degree to which findings of a study can be generalized to other populations Rationale (CORRECT): 2. External validity is concerned with generalization of study findings to other populations.

Non-probability sampling

(Convenience, quota, purposive) -Inclusion group is NOT random -LESS generalizable -LESS representative

Probability sampling

(Simple random sampling, stratified random, and multi-stage cluster) -Use RANDOMIZATION -MORE generalizable -MORE representative

Manipulation

-"Doing something" to at least some of the subjects -In IV is manipulated when some subjects (experimental groups) receive the intervention and others (control group) do not.

Convenience

-(Non-probability) -Use of the most readily accessible persons or objects as subjects in a study -Easy to recruit subjects -GREATEST RISK of BIAS in this sample -Used most with quantitative non-experimental or qualitative studies

Special sampling strategies (p.234)

-Non-probability -Matching -Networking sampling (snowballing)

Quota p.227

-Non-probability sampling-not randomly selected -Knowledge about characteristics of the population of interest is used to build representativeness into the sample. -Identifies the strata of the population and proportionally represents the strata in the sample. -

Components of feasibility

-Time -Subject availability -Facility equipment and availability -Expense -Experience -Ethics

3 parts to reliability

1. Stability: instrument's ability to produce the same results with repeated testing--Test-retest reliablity (Pearson r)(Parallel or alternate-form reliability) 2. Homogeneity: all of the items in an instrument measure the same concept or characteristic. Scale is unidimensional. (Item-to-total correlations, split-half reliability, KR-20 coefficient dichotomous response, Cronbach's alpha (likert-type response) 3. Equivalence: Produces the same result when equivalent or parallel instruments or procedures are used. Consistency or agreement among observers using the same instrument or between alternate forms of an instrument (Parallel or alternate form, inter-rater reliability)

Cause and effect relationships--3 criteria needed to infer causuality

1. The causal (independent) and effect (dependent) variables must be associated with each other 2. The cause must precede the effect 3. The relationship must not be explainable by another variable.

To be scientific, observations must fulfill the following 4 conditions:

1. The observations undertaken are consistent with the study's aim/objectives 2. There is a standardized and systematic plan for the observation and the recording of data. 3. All of the observations are checked and controlled. 4. The observations are r/t scientific concepts and theories.

Which would be considered open-ended questions? (Select all that apply.) Please describe your impression of your most recent hospital stay. How many times per year do you typically visit your family physician? Do you experience this symptom more than once per month? Compare the quality of care provided by a nurse practitioner with that provided by a physician.

1. This is an example of an open-ended question. 4. This is an example of an open-ended question

3 kinds of validity

1. content 2. criterion-related 3. construct

A percentile of 90 is exceeded by what percentage of the scores? 10% 45% 66% 90%

10% Rationale (CORRECT): 1. The percentile represents the percentage of cases a given score exceeds. In this case, a percentile of 90 exceeds all but 10% of the scores.

What type of score does a Z score of ?2 indicate? Too low to be caused by chance alone 2 standard deviations below the mean Skewed to the right 2 SD above the mean

2 standard deviations below the mean Rationale (CORRECT): 2. A Z score of -2 means that the observation is 2 standard deviations below the mean.

Which would be considered to be closed-ended questions? (Select all that apply.) Please describe your impression of your most recent hospital stay. How many times per year do you typically visit your family physician? Do you experience this symptom more than once per month? Compare the quality of care provided by a nurse practitioner with that provided by a physician.

2. This is an example of a closed-ended question. 3. This is an example of a closed-ended question.

What percentage of scores fall within 1 standard deviation (SD) of the mean? 0% 34% 68% 95%

68%

Respondents scored the following on a math test. What is the mode? 100 85 100 85 95 80 93 78 85 76 100 93 85 80

85 Rationale (CORRECT): 3. The mode is the most frequent score. In this example, the most frequent score (three people) was 85.

Which one of the following is a biological measurement? Pain rated on a scale of 1 to 10 A blood glucose level of 110 mg/100 mL Pitting edema in both ankles Blue-tinged lips and nail beds

A blood glucose level of 110 mg/100 mL Rationale (CORRECT): 2. Biological measurement involves the use of specialized equipment to determine physical and biological characteristics of subjects.

What type of problem is introduced by the response style known as social desirability? Interrater reliability Consistency in responses Accuracy of responses Precision of measurement

Accuracy of responses Rationale (CORRECT): 3. When participants respond in a way that they think is socially desirable, their responses will likely be inaccurate.

A random sample of U.S. hospitals is surveyed to discover their nurse-to-patient ratios. What is the sampling frame for this study? All nurses hired by hospitals All U.S. hospitals All patients of the sample hospitals Hospitals responding to the survey

All U.S. hospitals Rationale (CORRECT): 2. All U.S. hospitals will have an equal opportunity to be included in the study.

What is a population that meets the population criteria for a study and is available? A target population An accessible population A convenience sample A representative population

An accessible population

The statistical result of a research study is reported as "F = 3.59, df = 2, 48, p = 0.05." Which statistical test produces these results? Mean Analysis of variance Chi-square Pearson correlation

Analysis of variance Rationale (CORRECT): 2. This is an accurate description of how to report analysis of variance.

Internal validity

Asks whether the independent variable really made the difference of change in the dependent variable.

How can consistency in data collection be ensured? By using physiologic data-collection methods By having one person collect the data By avoiding the use of observation methods By collecting data from each subject in the same wa

By collecting data from each subject in the same way Rationale (CORRECT): 4. Consistency refers to the collection of data in the same way by all data collectors throughout the study.

Subjects should demonstrate homogeneity with respect to which variables? A. Independent B. Dependent C. Extraneous D. Outcome

C. Extraneous

To avoid threats to validity, a researcher must identify, plan for, and control: A. Independent variables. B. Unexpected findings. C. The control group. D. Extraneous variables.

D. Extraneous variables.

Exclusion criteria is also called

Deliminations

Which of the following types of studies is concerned specifically with a time perspective? Developmental studies Cross-sectional studies Descriptive studies Causal modeling studies

Developmental studies Rationale (CORRECT): 1. Investigators who use developmental studies are concerned with changes that result from elapsed time. Rationale: Cross-sectional studies collect data at only one point in time. Descriptive studies yield information about the characteristics of variables and do not necessarily have a specific time perspective. Causal modeling techniques establish predictive or causal links among variables and do not necessarily have a time perspective.

Randomization

Each subject in a population has an equal chance of being selected

A researcher decides to use six people to help him collect data for a quantitative study. Which one of the following is potentially a threat to the internal validity of this study? History Instrumentation Maturation Selection effects

Instrumentation: A bias could be introduced if the six people involved in data collection are not consistent.

What is the degree of consistency in data collection between two investigators called? Interrater reliability Reactivity Debriefing Cross-rater validity

Interrater reliability Rationale (CORRECT): 1. The degree to which two investigators collect data consistently is referred to as interrater reliability.

Which level of measurement uses the median? (Select all that apply.) Nominal Ordinal Interval Ratio

Interval measurement uses the mode, median, and mean. Rationale (CORRECT): 4. Ratio measurement uses the mode, median, and mean. Incorrect Feedback: Rationale (INCORRECT): 1. Nominal measurement uses the mode. Rationale (INCORRECT): 2. Ordinal measurement uses the mode and median.

A standard deviation: Is always reported with the mean Can be calculated from the mean Determines the shape of the normal curve Is a measure of central tendency

Is always reported with the mean Rationale (CORRECT): 1. SD is always reported with the mean. INCORRECT: It is the frequency distribution, not the SD, that determines the shape of a curve. Rationale (INCORRECT): 4. Measures of central tendency include mean, median, and mode.

Why are non-experimental correlational studies used frequently in nursing research? Findings of nonexperimental correlational studies can be generalized to larger populations. Independent variables can be manipulated very precisely in correlational studies. Many of the phenomena of clinical interest do not lend themselves to manipulation, control, or randomization. To determine the best clinical practices, nurses must be aware of cause-and-effect relationships.

Many of the phenomena of clinical interest do not lend themselves to manipulation, control, or randomization. Rationale (CORRECT): 3. Researchers cannot manipulate, control, or randomize many phenomena of concern to nurses.

The statistical result of a research study is reported as "m = 118.28." Which statistical test produces these results? Mean Analysis of variance Chi-square Pearson correlation

Mean Rationale (CORRECT): 1. This is an accurate description of how to report the mean.

Central tendency

Mean: average Median: one in the middle 50% above and 50% below Mode: one occurring most often

Sampling is found in what section?

Methods

What is the approach used when a variety of measurement strategies are used to examine the relationship between instruments that should measure the same construct and between those that should measure different constructs? Contrasted-groups approach Convergent approach Multitrait-multimethod approach Divergent approach

Multitrait-multimethod approach Rationale (CORRECT): 3. The multitrait-multimethod approach involves examining the relationship between instruments that should measure the same construct and between those that should measure different constructs. Rationale (INCORRECT): 1. In the contrasted-groups approach, the researcher administers an instrument to two groups of individuals expected to score extremely high and low in the characteristic being measured. Rationale (INCORRECT): 2. In the convergent approach, two or more tools that theoretically measure the same construct are identified and administered to the same subjects. Rationale (INCORRECT): 4. The divergent approach uses measurement approaches that differentiate one construct from others that may be similar.

Data collection methods must be objective and systematic. Define objective and systematic

Objective: free from the researcher's personal biases, beliefs, values, or attitudes Systematic: data are collected in a uniform, consistent or standard way

Data collection methods

Observational methods-unstructured, structured Self-report methods-interviews, questionnaires, demographic data Physiological measurement methods Existing data methods

systematic error

Occurs when scores are incorrect, but they are incorrect in the same direction. For example: when all subjects were assigned to a weight scale that is under by 3lbs for all the subjects in the study

The statistical result of a research study is reported as "r = 0.49, p = 0.01." Which statistical test produces these results? Mean Analysis of variance Chi-square Pearson correlation

Pearson correlation Rationale (CORRECT): 4. This is an accurate description of how to report Pearson correlation.

Positive/negative skew

Positive skew: when the mean is to the right of the median Negative skew: when the mean is to the left of the median

Which one of the following is true about both nonprobability sampling techniques and probability sampling techniques? Nonprobability sampling techniques are more rigorous. Each element has an equal chance of being included in the study with either technique. Probability sampling more likely results in a representative sample. Generalizability of findings is not possible with either technique.

Probability sampling more likely results in a representative sample.

What is another name for longitudinal studies? (Select all that apply. Prospective studies Cohort studies Repeated measures studies Ex post facto studies

Prospective studies Cohort studies Repeated measures studies

A researcher wants to study specific characteristics of adult triplets who were raised separately. Which of the following sampling techniques would the researcher most likely use? Convenience Quota Purposive Random

Purposive

Self-Report Methods

Questionnaires Paper and pencil instruments May be open-ended or closed-ended Referred to as instruments, surveys, scales or tools Most useful when there is a finite set of questions Likert-type Scale - closed-ended How satisfied are you with your current position?

Which sampling strategy is considered time-consuming in terms of drawing a sample? Convenience Quota Simple random Stratified random

Rationale (CORRECT): 4. Stratified random sampling is considered time consuming in terms of drawing a sample.

What is the most important characteristic of a sample? Delimitation Purposiveness Representativeness Accessibility

Representativeness

Observing subjects without their knowledge violates which human right? Fair treatment Right to information Freedom from risk of harm Right to informed consent

Right to informed consent Rationale (CORRECT): 4. Individuals have the right to choose whether or not to participate in a research study.

A parameter is a characteristic of a population; of what is a statistic characteristic? Normal distribution SD Sample Hypothesis

Sample

A researcher wanted to conduct a study of depression among men working in executive corporate positions. She located 10 subjects by advertising and then invited each of those subjects to refer their acquaintances who met the study criteria. This sampling technique is called: Snowballing Matching Clustering Systematic

Snowballing Rationale (CORRECT): 1. Snowballing refers to finding people who meet the study criteria through social networks. Rationale (INCORRECT): 2. Matching refers to constructing an equivalent comparison sample group by filling it with subjects who are similar to each subject in the control group. Rationale (INCORRECT): 3. Clustering refers to a multistage process of successive random sampling. Rationale (INCORRECT): 4. There is no sampling technique called systematic.

Considerations with self-report methods

Social desirability --There is no way to tell whether the respondent is telling the truth or responding in a socially desirable way --People are known to respond to questions in a way that makes a favorable impression Respondent burden --If the questionnaire length or interview is too long or questions are too difficult to answer in a reasonable amount of time considering age, health condition, or mental status or participants --May result in incomplete or erroneous answers as well as missing data

Divergent validity

Sometimes called discriminant validity uses measurement approaches that differentiate one construct from others that may be similar Established by analyzing the KPSC scores for the four groups with the one way analysis ANOVA and post hoc tests.

Operational definition

The process of translating a concept into a measurable variable for data collection

What is the relationship between the error variance, the true variance, and the observed score called? Parallel reliability Measurement error Homogeneity The reliability coefficient

The reliability coefficient Rationale (CORRECT): 4. The reliability coefficient measures the relationship among the error variance, the true variance, and the observed score.

Which of the following statements must apply to observations for them to be scientific? (Select all that apply.) They must be consistent with the study's objectives. They must be recorded in a standardized way. They must be conducted by one investigator. They must be focused on concepts that are difficult to measure objectively.

They must be consistent with the study's objectives. They must be recorded in a standardized wa

What is the purpose of establishing inclusion and exclusion criteria? To randomize sample selection To delineate an accessible population To control for potential bias To clarify the research question

To control for potential bias

When is bias introduced into a study? When sample selection is carried out inconsistently When subjects do not volunteer to participate When probability sampling techniques are used When the population is homogeneous

When sample selection is carried out inconsistently Rationale (CORRECT): 1. Inconsistent sample selection jeopardizes the representativeness of the sample and introduces bias. Rationale (INCORRECT): 2. Random sampling introduces the least amount of bias in the sample. Rationale (INCORRECT): 3. Probability sampling introduces the least amount of bias in the sample. Rationale (INCORRECT): 4. Homogeneity in the population reduces the likelihood of bias.

When is the t statistic is used? When the investigator is measuring the differences of means between two groups When the investigator is measuring the differences in frequencies of each category When the investigator is measuring the degree of association between two or more variables When the investigator is measuring the relationship between one interval-dependent variable and several independent variables

When the investigator is measuring the differences of means between two groups T for TWO Rationale (CORRECT): 1. The t statistic is used to measure the differences of means between two groups.

Validity refers to...

Whether a measurement instrument ACCURATELY measures what it is supposed to measure **High reliability may not be congruent with evidence of VALIDITY A measure can be reliable but NOT valid A valid instrument, however, is RELIABLE

A researcher studied the behavior of children who had witnessed the loss of a parent. She observed the children at play by watching through a one-way mirror. What is this type of observation called? Concealment without intervention Concealment with intervention No concealment without intervention No concealment with intervention

Concealment without intervention Rationale (CORRECT): 1. The researcher was concealed from the children and did not intervene with them.

The statistical result of a research study is reported as "X2 = 2.52, df = 1, P = 0.05." Which statistical test produces these results? Mean Analysis of variance Chi-square Pearson correlation

Chi-square

What do the data contain when a subject is answering a question in a socially desirable way? Error variance Random error Constant error Unreliability

Constant error Rationale (CORRECT): 3. Social desirability is a form of constant error. Validity is impacted with social desirability

Why is it imperative that data collection methods be objective? Data must be collected in the same way throughout the study. The researcher must clearly define each dependent variable. Data must not be influenced by the people collecting data. Every variable must be measurable.

Data must not be influenced by the people collecting data.

Nominal

Descriptive categories Only labels (names) Provides the least amount of mathematical possibilities Either dichotomous or categorical Ex: gender

What is the extent of variability in test scores that is attributable to error rather than a true measure of the dependent variable called? Chance error Error variance Constant error Stability

Error variance Rationale (CORRECT): 2. The error variance is the difference in scores between the actual measurement of behavioral changes and the extent of variability in test scores

Nonexperimental studies provide what type of evidence? Level II Level III Level IV Level V

Level IV

Which type of study is a case-control study? Survey Correlational Developmental Experimental

Developmental Rationale (CORRECT): 3. A case-control study is a developmental study.

Which type of study is a cohort study? Survey Correlational Developmental Experimental

Developmental Rationale (CORRECT): 3. A cohort study is a developmental study.

Which type of study is a longitudinal study? Survey Correlational Developmental Experimental

Developmental Rationale (CORRECT): 3. A longitudinal study is a developmental study.

Which type of study is a prospective study? Survey Correlational Developmental Experimental

Developmental Rationale (CORRECT): 3. A prospective study is a developmental study.

Which type of study is an ex post facto study? Survey Correlational Developmental Experimental

Developmental Rationale (CORRECT): 3. An ex post facto study is a developmental study.

The t statistic is used when the investigator is measuring the: Differences of means between two groups Differences in frequencies of each category Degree of association between two or more variables Relationship between one interval-dependent variable and several independent variables

Differences of means between two groups Rationale (CORRECT): 1. The t test statistic is used to measure the differences of means between two groups.

Researchers administered one tool that measured the concept of hope and another that measured the concept of anxiety to the same group of subjects. The scores on the first instrument were negatively related to the scores on the second instrument. This is a measure of what type of validity? Divergent validity Convergent validity Face validity Construct validity

Divergent validity Rationale (CORRECT): 1. Divergent validity uses measurement approaches that differentiate one construct from others that may be similar.

What is the role of an investigator in a correlational study? Examines the relationship between two or more variables. Evaluates data-collection instruments. Determines a cause-and-effect relationship among variables. Uses each subject as his or her own control

Examines the relationship between two or more variables. Rationale (CORRECT): 1. The investigator measures the degree to which variables covary.

Using a quota sampling technique introduces what type of bias? Internal validity External validity Face validity Content validity

External validity Rationale (CORRECT): 2. As with any nonprobability sample, the quota strategy contains an unknown source of bias that affects external validity.

All items in a tool should measure the same concept or characteristic. What is this called? Stability Homogeneity Equivalence Precision

Homogeneity Rationale (CORRECT): 2. Homogeneity means that all items in a tool measure the same concept or characteristic

What attribute of an instrument is the KR-20 coefficient measuring? Internal validity Equivalence Homogeneity Stability

Homogeneity Rationale (CORRECT): 3. The Kuder-Richardson (KR-20) coefficient is used to estimate the homogeneity of instruments with a dichotomous response format. Rationale (INCORRECT): 1. Item-to-total correlations measure the relationship between each of the items and the total scale. Rationale (INCORRECT): 3. Cronbach's alpha measures the homogeneity of an instrument with a Likert-type format. Rationale (INCORRECT): 4. The KR-20 coefficient is used to estimate the homogeneity of instruments with a dichotomous response format.

random error

Occurs when scores vary in a random way, and it occurs when data collectors do not use standard procedures to collect data consistently among all subjects in a study.

Respondent burden

Occurs when the length of the questionnaire or interview is too long or the questions too difficult for respondents to answer in a reasonable amount of time considering their age, health condition, or mental status. Can result in incomplete or erroneous answers, as well as missing data, which may jeopardize the validity of the study findings.

Which reliability measure is considered to be a measure of both stability and equivalence? (Select all that apply). Item-to-total correlation Split-half reliability Interrater reliability Parallel, or alternate-form, reliability Test-retest reliability Kuder-Richardson coefficient Cronbach's alpha

Parallel, or alternate-form, reliability is considered a test of both stability and equivalence.

How is control accomplished?

-Accomplished by ruling out extraneous or mediating variables that compete with the independent variable as an explanation for a study's outcome.

Which type of study is not only concerned with the existing status and the relationship and differences among phenomena at one point in time but also with changes that result from elapsed time?

-Developmental studies

Reliability coefficient

-Expresses the relationship between the error variance, true variance, and the observed score Scored between 0 and 1 Closer to 1 the more reliable!!!!! Instrument should be 0.70 or higher to be considered reliable

No concealment without intervention

-Informed consent is obtained from the subject, and then behavior is observed

The types of samples

-Non-probability sampling: convenience, quota, purposive -Probability sampling: simple random sampling, stratified random sampling, multi-stage (cluster) sampling

Secondary analysis

-Not a design but a research method that takes previously collected and analyzed data from an original study and reanalyzes the data for a second purpose. -**cannot be used to infer causality--only associations or relationships.

No concealment with intervention

-Observing the effects of an intervention introduced for scientific purposes -Subjects know they are participating in a research study -Reactivity is a problem with this type of study

Multistage (cluster) sampling

-Probability sampling -A successive random sampling of units (clusters) that progress from large to small and meet sample eligibility criteria -Sampling units or clusters that can be selected by simple random or stratified random sampling methods -ex: Large national survey stage 1=states stage 2=cities and stage 3=households Advantages: more economical in terms of time & money Disadvantages: More sampling errors tend to occur than with simple random or stratified random sampling

Simple Random Sampling

-Researcher defines the population (a set), lists all of the units of the population (a sampling frame), and selects a sample of units (a subset) from which the sample will be chosen. -The set of example could be all nurse's with a PhD; the sampling frame then would be a list of those nurses. A random sample can then be chosen from the list by using a computer program that generates the order of selection from the list.

Advantages of Simple Random Sampling

-Sample selection is not subject to conscious bias -Representativeness is maximized -Differences in the characteristics of the sample and the population are purely a function of chance -Probability of choosing a non-representative sample decreases as the size of the sample increases.

Power analysis

-Sample size is calculated with power analysis -G-power: free software to calculate sample size for quantitative study

Concealment with intervention

-Staging a situation and observing the behaviors that are evoked in the subjects as a result of the intervention -Because the subjects are unaware of their participation, this type of observation has fallen into disfavor and rarely is used in nursing research

Concealment

-Subjects do not know they are being observed. -Used if there is concern that a subject's behavior will change as a result of being observed (reactivity) -Debriefing is used to inform subjects after the observation, allowing them the opportunity to refuse to have their data included in the study and discussing any questions they might have

Stratified Random Sampling

-The population is divided into strata of subgroups that are homogenous -It allows more representativeness

Concealment without intervention

-The researcher watches the subjects without their knowledge of the observation and does not provoke the subjects into action -Use hidden television cameras, audiotapes, or one-way mirrors; this method is often used in observational studies of children

Disadvantages of Simple Random Sampling

-Time consuming and usually inefficient method of obtaining a random sample

Self-Report Methods

-Use interviews or structured questionnaires to collect information about experiences, feelings, behaviors, or attitudes -Common in nursing research and most useful for collecting data on variables that cannot be directly observed or measured -Like quality of life, satisfaction with nursing care, social support, uncertainty, and functional status

Reliability is

-the ability of an instrument to measure the attributes of a concept or construct CONSISTENTLY -the extent to which the instrument yields the same results on repeated measures.

Participants in a research study were questioned about their satisfaction with seven aspects of their education. They were asked to rate the degree of their satisfaction on a scale of 1 (extremely satisfied) to 5 (extremely unsatisfied). This response format is an example of: Dichotomous variables A Likert scale Nominal data Open-ended responses

A Likert scale Rationale (CORRECT): 2. A response format using a scale of 1 to 5 is called a Likert scale.

Graduates of nursing programs were surveyed 6 months postgraduation to determine their perspectives on the adequacy of their nursing education. What type of study is this an example of? A cross-sectional study A prospective study An ex post facto study A longitudinal study

A cross-sectional study Rationale (CORRECT): 1. Data were collected at one point in time with the same subjects. Prospective studies collect data at various times for a period of time. Rationale (INCORRECT): 3. Ex post facto studies collect data retrospectively. Rationale (INCORRECT): 4. Longitudinal studies collect data retrospectively.

A Pearson r coefficient of 0.75 indicates: Little association between ordinal variables A great deal of association between interval level measurements No association between ratio level measurements Some association between nominal level measurements

A great deal of association between interval level measurements Rationale (CORRECT): 2. The Pearson r coefficient measures the association between interval level measurements.

A Cronbach's alpha coefficient of 0.94 was obtained for an instrument. What does this indicate? A low degree of validity A high degree of internal consistency A low degree of homogeneity A high degree of stability

A high degree of internal consistency Rationale (CORRECT): 2.The higher the coefficient, the higher the degree of internal consistency. Rationale (INCORRECT): 1. Cronbach's alpha tests for reliability, not validity. Rationale (INCORRECT): 3. The higher the coefficient, the higher the degree of homogeneity. Rationale (INCORRECT): 4. Cronbach's alpha does not test for the stability of an instrument.

Which of the following is an example of a primary source? A published commentary on the findings of another study A doctoral dissertation that critiques all research in the area of attention deficit disorder A textbook of medical-surgical nursing A journal article about a study that used large, previously unpublished databases generated by the U.S. census

A journal article about a study that used large, previously unpublished databases generated by the U.S. census

Which of the following statements are true? (Select all that apply.) A measure can be valid but not reliable. A measure can be reliable but not valid. A measure can be neither reliable nor valid. A measure can be both reliable and valid.

A measure can be reliable but not valid. A measure can be neither reliable nor valid. A measure can be both reliable and valid.

How is interrater reliability expressed? The total number of differences among scorers subtracted from the total number of items The total number of differences among scores The average score among scorers A percentage of agreement between scores

A percentage of agreement between scores Rationale (CORRECT): 4. When two or more investigators are collecting data, interrater reliability is determined by calculating the percentage of agreement between their scores.

What is an advantage of using surveys for data collection? Causation can be determined. Information can be collected in great depth. Large-scale surveys are relatively inexpensive. A relatively small number of subjects can provide accurate information.

A relatively small number of subjects can provide accurate information. Rationale (CORRECT): 4. If a sample is representative of the population, a relatively small number of subjects can provide an accurate picture of the population.

A nurse conducted a research study that compared patient outcomes (length of hospital stay and perceived knowledge of disease process) in groups of 12-year-old patients with a diagnosis of epilepsy who were being treated at St. Joseph's Hospital. Which of the following are delimitations of this study? (Select all that apply). Gender Age Socioeconomic status Diagnosis

Age & Diagnosis

Anecdotes

Are summaries of a particular observation that usually focus on the behaviors of interest and frequently add to the richness of research reports by illustrating a particular point.

Which critiquing criterion would apply to records and databases? Is there a provision for evaluating the accuracy of the instrument? Are the majority of the items appropriately close-ended or open-ended? Is the interview schedule described adequately enough to know whether it covers the topic? Are the existing data used appropriately considering the problem being studied?

Are the existing data used appropriately considering the problem being studied?

Which critiquing criterion would apply to questionnaires? Is there a provision for evaluating the accuracy of the instrument? Are the majority of the items appropriately close-ended or open-ended? Is the interview schedule described adequately enough to know whether it covers the topic? Are the existing data used appropriately considering the problem being studied?

Are the majority of the items appropriately close-ended or open-ended?

One way to determine whether findings from one study can be generalized to the larger population is to: Determine a Pearson r coefficient Calculate a confidence interval Examine the research findings for bias Match findings from multiple studies that have similar samples

Calculate a confidence interval Rationale (CORRECT): 2. A confidence interval quantifies the uncertainty of a statistic or the probable value range within which a population parameter is expected to lie. Rationale (INCORRECT): 1. A Pearson r coefficient measures the degree of association between two or more variables. Rationale (INCORRECT): 3. Although bias reduces the generalizability of research findings, there is a more direct approach to assessing the generalizability of research findings to the larger population; this is not the best answer. Rationale (INCORRECT): 4. Findings from multiple studies with similar samples may or may not be generalizable to the larger population; there is a better answer to this question.

Reliability is concerned with which type of error?

Chance/random error

Which of the following is true about the chi-square statistic? Chi-square is used when the sample is small and expected frequencies are less than 6 in each cell. Chi-square is used to measure the degree of association between two or more variables. Chi-square is used to determine whether the frequency in each category is different from what would be expected by chance. Chi-square measures whether two group means are different.

Chi-square is used to determine whether the frequency in each category is different from what would be expected by chance. Rationale (CORRECT): 3. Chi-square is a nonparametric statistic used to determine whether the frequency in each category is different from what would be expected by chance. Rationale (INCORRECT): 1. Fisher's exact probability test is used in these conditions. Rationale (INCORRECT): 2. Correlations are used to measure the degree of association between two or more variables. Rationale (INCORRECT): 4. The t statistic is used to measure whether two group means are different.

Which measurement would a researcher use to test for reliability when the data are in Likert-scale response format? Chronbach's alpha KR-20 coefficient Parallel, or alternate-form, reliability Split-half reliability

Chronbach's alpha Rationale (CORRECT): 1. Cronbach's alpha measures the homogeneity of an instrument with a Likert-type format.

To conduct a practice analysis of nurse aides, a researcher selected a sample of nurse aides by first obtaining a random sample of all hospitals in the United States, then acquiring a list of every aide working in those hospitals, and finally randomly selecting two aides from each hospital. This sampling technique is called: Quota sampling Stratified random sampling Purposive sampling Cluster sampling

Cluster sampling Rationale (CORRECT): 4. The sample for this study was chosen by "clustering" each level of sample selection. Rationale (INCORRECT): 1. In quota sampling, subjects are hand-selected to ensure a proportional number of subjects with selected characteristics. Rationale (INCORRECT): 2. In stratified random sampling, the population is divided into strata, or subgroups, before random selection of subjects. Rationale (INCORRECT): 3. A purposive sample is used when a highly unusual group is being studied.

The nurse researcher has decided to determine factors predictive of parenting self-efficacy at 12 to 48 hours after childbirth and at 1 month postpartum. Which design of types would be applicable? (Select all that apply). Cohort Descriptive Longitudinal with retrospective Prospective

Cohort and prospective

A study was designed to assess whether motion had an effect on the mood of elderly residents in a nursing home. The researcher examined whether residents who ambulated for an hour every day were more engaged in their environments than those who did not ambulate. This is an example of which type of study? Prospective Comparative Ex post facto Correlational

Comparative Rationale (CORRECT): 2. Comparative studies determine differences between variables.

Researchers used an instrument to measure self-esteem in adolescent mothers. To measure the validity of this instrument, they used a second instrument known to measure self-esteem in women. What type of validity were they measuring? Predictive validity Face validity Concurrent validity Construct validity

Concurrent validity Rationale (CORRECT): 3. Concurrent validity refers to the degree of correlation of two measures of the same concept administered at the same time.

What does validity concern? Instability Random error Constant error Chance error

Constant error Rationale (CORRECT): 3. Constant error impacts the validity of a study. Instability impacts the reliability of a study. Rationale (INCORRECT): 2. Random error impacts the reliability of a study. Rationale (INCORRECT): 4. Chance error impacts the reliability of a study.

Sometimes two or more tools that theoretically measure the same construct are administered to the same subjects and a correlational analysis is performed. This is a strategy to test what type of validity? Predictive validity Face validity Concurrent validity Construct validity

Construct validity Rationale (CORRECT): 4. Construct validity refers to the extent to which a test measures a theoretical construct or trait.

Data analysis for unstructured response formats requires what type of method? Frequency distribution Interrater agreement Averaging of responses Content analysis

Content analysis Rationale (CORRECT): 4. Content analysis allows the analysis of data that are unstructured.

A researcher identifies two groups of individuals: one expected to score extremely high and the other expected to score extremely low on an instrument measuring the construct "empowerment." The instrument is administered, and the differences in scores are examined. Of the following, what are the approach being used and the type of validity being measured? Contrasted-groups approach to test construct validity Divergent approach to test criterion-related validity Multitrait-multimethod approach to test construct validity Convergent approach to test face validity

Contrasted-groups approach to test construct validity Rationale (CORRECT): 1. In the contrasted-groups approach, the researcher administers an instrument to two groups of individuals expected to score extremely high and low in the characteristic being measured. Construct validity is being measured.

Which sampling strategy is considered to be very easy in terms of drawing a sample? Convenience Quota Simple random Stratified random

Convenience

Which sampling strategy is considered to have the greatest risk of bias of any strategy? Convenience Quota Simple random Stratified random

Convenience

n a nationwide study, a researcher collected blood samples from patients who had been diagnosed with clinical depression. She also collected blood samples from each patient's nuclear family. Which sample technique did she most likely use? Purposive Convenience Quota Random

Convenience Rationale (CORRECT): 2. Because the population of patients diagnosed with clinical depression is not known, quota sampling and random sampling were not possible. There were few selection criteria, making purposive sampling unnecessary. The easiest and most commonly used sampling technique is convenience sampling. Rationale (INCORRECT): 1. Because the total population of patients with clinical depression is not known, handpicking a "typical" sample would be impossible. Rationale (INCORRECT): 3. Because the total population of patients with clinical depression is not known, handpicking a representative sample would be impossible. Rationale (INCORRECT): 4. Because the population of patients diagnosed with clinical depression is not known, random sampling was not possible.

Voluntary participation in a study may result in a sample that feels strongly about the issue being studied. This is an issue in which type of sampling method? Convenience sampling Probability sampling Stratified random sampling Cluster sampling

Convenience sampling Rationale (CORRECT): 1. When a researcher selects a sample based on convenience, the individuals willing to participate may do so because of their strong feelings about the topic being studied. They might not be representative of the larger population.

Which of the following statements are true about convenience sampling? (Select all that apply.) Convenience sampling is the most commonly used method of sampling. Convenience sampling is the strongest nonprobability sampling strategy. Research consumers should be skeptical about the external validity of the finding. Convenience sampling provides level III evidence.

Convenience sampling is the most commonly used method of sampling. Research consumers should be skeptical about the external validity of the finding.

Which one of the following levels of measurement has a scale with an absolute zero? Ratio Nominal Interval Ordinal

Correct Ratio Rationale (CORRECT): 1. Ratio data are rankings of variables on scales with an absolute zero.

Which study would be considered a developmental study? Descriptive Correlational Developmental Exploratory Comparative Cross-sectional

Cross-sectional Rationale (INCORRECT): 1. A descriptive study could be considered to be a survey. Rationale (INCORRECT): 2. A correlational study would be a relationship-difference study. Rationale (INCORRECT): 3. A developmental study would be a relationship/difference study. Rationale (INCORRECT): 4. An exploratory study could be considered a survey. Rationale (INCORRECT): 5. A comparative study could be considered a survey.

Constancy

Data collection procedures are the same for each subject; data collected in the same manner and under the same conditions. -Training and supervising individuals delivering interventions and collecting data--constancy and intervention fidelity

What is a process that protects the rights of subjects who have been observed without their knowledge? Reactivity Debriefing Content analysis Concealment.

Debriefing Rationale (CORRECT): 2. Debriefing is a process of informing subjects that they have been observed as part of a research study.

The nurse researcher has decided to assess patient and provider responses to a computerized symptom assessment system. Which design of types would be applicable? (Select all that apply). Exploratory Descriptive Longitudinal with retrospective Longitudinal medical records review

Descriptive Longitudinal with retrospective Longitudinal medical records review

A researcher explored the nursing practice of new graduates to measure the frequency of medical errors they made over time. A cohort of new graduates was surveyed at 6 months, 9 months, and 1 year postgraduation. Which of the following designs best describes this study? Correlational predictive Retrospective correlational Descriptive longitudinal Comparative descriptive

Descriptive longitudinal Rationale (CORRECT): 3. The study is descriptive because the researcher is measuring characteristics of the variable "frequency of medical errors" and longitudinal because data are being collected from one cohort of subjects repeatedly over time.

When does selection of data-collection methods occur? During the literature review From data collection In sample selection During the review of protection of human subj

During the literature review 1. During the review of literature, the researcher begins to explore how previous investigators defined and operationalized variables similar to those of interest in the current study.

What does a test score or measurement consist of? Random errors plus constant errors Random errors plus the true score Constant errors plus error variance Error variance plus the true score

Error variance plus the true score Rationale (CORRECT): 4. A test score includes the inherent bias of the study (error variance) and the true measurement.

Which critical appraisal question would be categorized as pertaining to research question and hypothesis? Given the level of evidence suggested by the research question, hypothesis, and design, what is the potential applicability to practice? Is there a synthesis summary that presents the overall strengths and weaknesses and arrives at a logical conclusion that generates hypotheses or research questions? Would it be possible to replicate this study in another clinical setting? What are the sources of bias and are they dealt with appropriately?

Given the level of evidence suggested by the research question, hypothesis, and design, what is the potential applicability to practice? Rationale (CORRECT): 1. This would be most accurately categorized as research question and hypothesis.

Which critical appraisal questions would be applicable to data collection methods and procedures? (Select all that apply). -How have the rights of subjects been protected? -Were the descriptive or inferential statistics appropriate to the level of measurement for each variable? -Did testing of the hypothesis(es) or research question(s) clearly support or not support each hypothesis or research question? -What are the limitations in generalizability? -Is a rationale given for why a particular instrument or method was selected? IncorrectTo what population may the findings be generalized? -Are the inferential statistics appropriate for the type of design, hypothesis(es) or research question(s)? -What provision is made for maintaining accuracy of the instrument and its use, if any?

Is a rationale given for why a particular instrument or method was selected? What provision is made for maintaining accuracy of the instrument and its use, if any?

An independent variable: Is the variable that is predicted to change Varies with a change in the dependent variable Is manipulated by the researcher Can only be identified by changes in the dependent variable

Is manipulated by the researcher Rationale (CORRECT): 3. The independent variable is manipulated by the researcher and has a presumed effect on the dependent variable.

Which critiquing criterion would apply to interviews? Is there a provision for evaluating the accuracy of the instrument? Are the majority of the items appropriately close-ended or open-ended? Is the interview schedule described adequately enough to know whether it covers the topic? Are the existing data used appropriately considering the problem being studied?

Is the interview schedule described adequately enough to know whether it covers the topic?

Which critiquing criterion would apply to physiologic measurement? Is there a provision for evaluating the accuracy of the instrument? Are the majority of the items appropriately close-ended or open-ended? Is the interview schedule described adequately enough to know whether it covers the topic? Are the existing data used appropriately considering the problem being studied?

Is there a provision for evaluating the accuracy of the instrument?

Which critical appraisal question would be categorized as pertaining to review of the literature? Given the level of evidence suggested by the research question, hypothesis, and design, what is the potential applicability to practice? Is there a synthesis summary that presents the overall strengths and weaknesses and arrives at a logical conclusion that generates hypotheses or research questions? Would it be possible to replicate this study in another clinical setting? What are the sources of bias and are they dealt with appropriately?

Is there a synthesis summary that presents the overall strengths and weaknesses and arrives at a logical conclusion that generates hypotheses or research questions? 2. This would most accurately categorize review of the literature.

Which critiquing criterion would apply to interviews? Were the observers trained to minimize any bias? Are operational definitions provided and clear? Is there clear indication that the subjects understood the task and the questions? Is there evidence that subjects were able to answer the questions?

Is there clear indication that the subjects understood the task and the questions?

Which critiquing criterion would apply to questionnaires? Were the observers trained to minimize any bias? Are operational definitions provided and clear? Is there clear indication that the subjects understood the task and the questions? Is there evidence that subjects were able to answer the questions?

Is there evidence that subjects were able to answer the questions?

What does a frequency distribution do? It determines the mean of the distribution. It determines the symmetry of the distribution. It demonstrates the measure of variability. It organizes data by the frequency of response.

It organizes data by the frequency of response. Rationale (CORRECT): 4. A frequency distribution groups the number of times each response is given.

Which measurement would a researcher use to test for reliability when the data are in dichotomous ("yes/no") format? Cronbach's alpha KR-20 coefficient Parallel, or alternate-form, reliability Split-half reliability

KR-20 coefficient Rationale (CORRECT): 2. KR-20 coefficient is used to estimate the homogeneity of instruments with a dichotomous response format.

Which measurement would a researcher use to test for reliability when the data are in dichotomous ("yes/no") format? Chronbach's alpha KR-20 coefficient Parallel, or alternate-form, reliability Split-half reliability

KR-20 coefficient Rationale (CORRECT): 2. The KR-20 coefficient is used to estimate the homogeneity of instruments. Rationale (INCORRECT): 1. Cronbach's alpha measures the homogeneity of an instrument with a Likert-type format. Rationale (INCORRECT): 3. Parallel, or alternate-form, reliability is an approach to measuring the stability of an instrument, regardless of its format. Rationale (INCORRECT): 4. Split-half reliability is an approach to test for the reliability of an instrument, regardless of its format.

What type of evidence is provided by meta-analysis? Level I Level II Level III Level IV

Level I

Non-experimental studies yield what level of evidence?

Level IV

Why are nonexperimental correlational studies used frequently in nursing research? Findings of nonexperimental correlational studies can be generalized to larger populations. Independent variables can be manipulated very precisely in correlational studies. Many of the phenomena of clinical interest do not lend themselves to manipulation, control, or randomization. To determine the best clinical practices, nurses must be aware of cause-and-effect relationships.

Many of the phenomena of clinical interest do not lend themselves to manipulation, control, or randomization. Rationale (CORRECT): 3. Researchers cannot manipulate, control, or randomize many phenomena of concern to nurses.

Parametric tests of significance are used when (select all that apply): Measurement is on an interval or ratio scale. At least one population parameter is estimated. The sample is small. Normal distribution cannot be assumed.

Measurement is on an interval or ratio scale. At least one population parameter is estimated.

When are parametric tests of significance used? (Select all that apply.) Measurement is on an interval or ratio scale. At least one population parameter is estimated. The sample is small. Normal distribution cannot be assumed.

Measurement is on an interval or ratio scale. At least one population parameter is estimated.

When data are skewed and the researcher is interested in the "typical" score, what is the best measure of central tendency to use? Median Mode Mean Normal curve

Median Rationale (CORRECT): 1. The median is not sensitive to extremes in high and low scores.

A research study explored the meaning of health for migrant farm-worker women in a southern state. A convenience sample of 20 migrant farm-worker women was surveyed. What is the population of this study? Migrant farm-worker women in Georgia Migrant farm workers Migrant farm-worker women Migrant farm workers in Georgia.

Migrant farm-worker women in Georgia

What is it called when subjects drop out of the study before completion of data collection? History Maturation Selection bias Mortality

Mortality

What is the best way to describe the distribution of the ages of all people who died in 2004? Normal distribution Positively skewed Negatively skewed Symmetrically distributed

Negatively skewed Rationale (CORRECT): 3. Age of death in the United States has a negative skew because most deaths occur at older ages.

A researcher attended an infant exercise program and observed mothers and infants interacting. What is this type of observation called? Concealment without intervention Concealment with intervention No concealment without intervention No concealment with intervention

No concealment without intervention Rationale (CORRECT): 3. The researcher was not concealed but did not intervene with the subjects.

In one study, the KR-20 coefficient was 0.01. What did this indicate? No correlation between the error variance, true variance, and observed score An almost perfect correlation between the error variance, true variance, and observed score Low measurement error An equal balance between the error variance and true variance

No correlation between the error variance, true variance, and observed score Rationale (CORRECT): 1. The lower the coefficient, the lower the correlation.

In one study, the Kuder-Richardson (KR-20) coefficient was 0.01. This indicated: No correlation was found between the error variance, true variance, and observed score. An almost perfect correlation was found between the error variance, true variance, and observed score. There was a low measurement error. An equal balance was found between the error variance and true variance.

No correlation was found between the error variance, true variance, and observed score. Rationale (CORRECT): 1. The lower the coefficient, the lower the correlation.

Purposive sampling

Non-probability sampling -Increasingly common strategy in which the researcher's knowledge of the population and its element is used to HANDPICK the cases to be included in the sample. -The researcher usually selects subjects who are considered to be typical of the population. -Useful in studying populations with unusual/rare characteristics -Assumes that errors of judgement in over or under representing characteristics of the population in the sample will tend to balance out.

Which of the following statements are true about nonexperimental designs? In nonexperimental studies, the independent variable is manipulated. Nonexperimental designs are used to test relationships among variables. Nonexperimental designs can be used to construct a picture of a phenomenon at one point in time. In nonexperimental studies, subjects are randomly selected.

Nonexperimental designs are used to test relationships among variables. Nonexperimental designs can be used to construct a picture of a phenomenon at one point in time.

What does the mean of a sampling distribution of means refer to? (Select all that apply.) Percentile Population Standard deviation Normal curve

Normal curve Population Rationale (INCORRECT): 1. Percentile refers to the distribution of scores about the mean. Rationale (INCORRECT): 3. Standard deviation refers to the difference between a score and the mean of scores.

What occurs when differences between scores are too large to be caused by chance? Null hypothesis is rejected. Scientific hypothesis is proven to be true. Intervention did not make a difference. Scientific hypothesi

Null hypothesis is rejected. Rationale (CORRECT): 1. The null hypothesis (the hypothesis that there are no differences) is rejected when there are differences between variables.

When a researcher does not know all of the possible alternative responses to a question, what type of question should be asked? Open-ended Close-ended Fixed-response Likert scale

Open-ended Rationale (CORRECT): 1. Open-ended questions allow the investigator to explore the scope of possible responses to a question. Rationale (INCORRECT): 2. Close-ended questions are best used when the scope of possible responses is known. Rationale (INCORRECT): 3. Fixed-response items are best used when the scope of possible responses is known. Rationale (INCORRECT): 4. A Likert scale is best used when the investigator wants to measure the strength of opinions or beliefs.

Which level of measurement focuses on relative rankings? Nominal Ordinal Interval Ratio

Ordinal Rationale (CORRECT): 2. Ordinal measurement focuses on relative rankings. Rationale (INCORRECT): 1. Nominal measurement focuses on classification. Rationale (INCORRECT): 3. Interval measurement focuses on rank ordering with equal intervals. Rationale (INCORRECT): 4. Ratio measurement focuses on rank ordering with equal intervals and absolute zero.

Which of the following levels of measurement are ranked variables or events with unequal intervals between numbers on the scale? Ratio Nominal Interval Ordinal

Ordinal Rationale (CORRECT): 4. Ordinal data are rankings of variables or events with unequal intervals.

A frequency distribution: Determines the mean of the distribution Determines the symmetry of the distribution Demonstrates the measure of variability Organizes data by the frequency of respo

Organizes data by the frequency of response

The institutional review board (IRB) is responsible for: Ensuring that the research design is appropriate Protecting subjects from undue risk Monitoring implementation of the study Overseeing sample selection

Protecting subjects from undue risk Rationale (CORRECT): 2. The purpose of an IRB is to oversee the protection of human subjects.

Which sampling strategy contains an unknown source of bias that affects external validity? Convenience Quota Simple random Stratified random

Quota Rationale (CORRECT): 2. Quota sampling contains an unknown source of bias that affects external validity. The very nature of a "quota" implies the selection of subjective boundaries for the sample.

What level of evidence do randomized clinical trials provide? II IV VI VII

Randomized clinical trials, or experimental studies, provide Level II evidence.

Which level of measurement focuses on rank ordering with equal intervals and absolute zero? Nominal Ordinal Interval Ratio

Ratio

Which studies would be considered survey studies? (Select all that apply). Descriptive Correlational Developmental Exploratory Comparative Cross-sectional

Rationale (CORRECT): 1. A descriptive study could be considered to be a survey. Rationale(CORRECT): 4. An exploratory study could be considered a survey. Rationale(CORRECT): 5. A comparative study could be considered a survey. Incorrect Feedback: Rationale (INCORRECT): 2. A correlational study would be a relationship-difference study. Rationale(INCORRECT): 3. A developmental study would be a relationship-difference study. Rationale(INCORRECT): 6. A cross-sectional study would be a developmental, relationship-difference study.

Which sampling strategy is considered relatively easy in terms of drawing a sample? (Select all that apply). Convenience Quota Simple random Stratified random

Rationale (CORRECT): 1. Convenience sampling is considered easy in terms of drawing a sample. Rationale (CORRECT): 2. Quota sampling is considered relatively easy in terms of drawing a sample. Incorrect Feedback: Rationale (INCORRECT): 3. Simple random sampling is considered laborious in terms of drawing a sample. Rationale (INCORRECT): 4. Stratified random sampling is considered time consuming in terms of drawing a sample.

Which reliability measures are considered to be tests of homogeneity? (Select all that apply). Item-to-total correlation Split-half reliability Interrater reliability Parallel, or alternate-form, reliability Test-retest reliability Kuder-Richardson coefficient Cronbach's alpha

Rationale (CORRECT): 1. Item-to-total correlation is considered a test of homogeneity. Rationale (CORRECT): 2. Split-half reliability is considered a test of homogeneity. Rationale (CORRECT): 6. Kuder-Richardson coefficient is considered a test of homogeneity. Rationale (CORRECT): 7. Cronbach's alpha is considered a test of homogeneity.

Which reliability measure is considered to be a test of homogeneity? (Select all that apply). Split-half reliability Test-retest reliability Parallel, or alternate-form, reliability Cronbach's alpha

Rationale (CORRECT): 1. Split-half reliability is considered a test of homogeneity. Rationale (CORRECT): 4. Cronbach's alpha is considered a test of homogeneity. Incorrect Feedback: Rationale (INCORRECT): 2. Test-retest reliability is considered a test of stability. Rationale (INCORRECT): 3. Parallel, or alternate-form, reliability is considered a test of equivalence.

Tests that are used to estimate the stability of an instrument include (select all that apply): Test-retest reliability. Kuder-Richardson (KR-20) coefficient. Parallel reliability. Cronbach's alpha

Rationale (CORRECT): 1. Test-retest measures the stability of an instrument. Rationale (CORRECT): 3. Parallel reliability measures the stability of an instrument.

Which tests are used to estimate the stability of an instrument? (Select all that apply): Test-retest reliability Kuder-Richardson (KR-20) coefficient Parallel reliability Cronbach's alpha

Rationale (CORRECT): 1. Test-retest measures the stability of an instrument. Rationale (CORRECT): 3. Parallel reliability measures the stability of an instrument. Incorrect Feedback: Rationale (INCORRECT): 4. Cronbach's alpha measures the homogeneity of an instrument. Rationale (INCORRECT): 2.The Kuder-Richardson (KR-20) coefficient measures the homogeneity of an instrument.

Which of the following conditions enhance the generalizability of study findings? (Select all that apply.) Few threats to validity Small sample size Random selection of subjects A representative sample

Rationale (CORRECT): 1. The fewer the threats to validity, the more likely the findings can be generalized to the larger population. Rationale (CORRECT): 3. Random selection of subjects enhances the generalizability of study findings. Rationale (CORRECT): 4. A sample that is representative of the larger population enhances the generalizability of study findings. Incorrect Feedback:

Which of the following statements are true? (Select all that apply.) Information found in tables should not be included in the text. Nonsupported data should be included in research reports in the same detail as the supported data. Significant findings establish a study's merit. Serendipitous findings should not be included in the results section.

Rationale (CORRECT): 1. The information found in tables should not be repetitious of the text. Rationale (CORRECT): 2. Nonsupported data can be as informative as supported data. Incorrect Feedback: Rationale (INCORRECT): 3. Studies with nonsignificant findings can be as valuable as those with significant findings. Rationale (INCORRECT): 4. Serendipitous findings should be included in the results section.

Which statements are applicable to type II errors? (Select all that apply.) They occur when the sample size is too small. Mean rejects the null hypothesis when it is actually true. Mean accepts the null hypothesis when it is false. It can be prevented by setting the level of significance before the study begins.

Rationale (CORRECT): 1. Type II errors (acceptance of a null hypothesis that is actually false) occur when the sample is too small, thereby limiting the opportunity to measure the treatment effect. Rationale (CORRECT): 3. This is a type II error.

Which study would be considered a relationship-difference study? (Select all that apply). Descriptive Correlational Developmental Exploratory Comparative Cross-sectional

Rationale (CORRECT): 2. A correlational study would be a relationship-difference study. Rationale (CORRECT): 3. A developmental study would be a relationship-difference study. Rationale (CORRECT): 6. A cross-sectional study would be a developmental, relationship-difference study. Incorrect Feedback: Rationale (INCORRECT): 1. A descriptive study could be considered a survey. Rationale (INCORRECT): 4. An exploratory study could be considered a survey. Rationale (INCORRECT): 5. A comparative study could be considered a survey.

Which of the following statements are true about nonexperimental designs? (Select all that apply.) In nonexperimental studies, the independent variable is manipulated. Nonexperimental designs are used to test relationships among variables. Nonexperimental designs can be used to construct a picture of a phenomenon at one point in time. In nonexperimental studies, subjects are randomly selected.

Rationale (CORRECT): 2. Nonexperimental designs are used in studies in which the researcher wishes to test relationships among variables. Rationale (CORRECT): 3. Nonexperimental designs are used in studies in which the researcher wishes to construct a picture of a phenomenon.

Which reliability measures are considered to be tests of equivalence? (Select all that apply). Item-to-total correlation Split-half reliability Interrater reliability Parallel, or alternate form, reliability Test-retest reliability Kuder-Richardson coefficient Cronbach's alpha

Rationale (CORRECT): 3. Interrater reliability is considered a test of equivalence. Rationale (CORRECT): 4. Parallel, or alternate-form, reliability is considered a test of both stability and equivalence.

Which level of measurement focuses on rank ordering with equal intervals? (Select all that apply). Nominal Ordinal Interval Ratio

Rationale (CORRECT): 3. Interval measurement focuses on rank ordering with equal intervals Rationale (CORRECT): 4. Ratio measurement focuses on rank ordering with equal intervals and absolute zero. Incorrect Feedback:

Which sampling strategy is considered laborious in terms of drawing a sample? (Select all that apply). Convenience Quota Simple random Stratified random

Rationale (CORRECT): 3. Simple random sampling is considered laborious in terms of drawing a sample. Rationale (CORRECT): 4. Stratified random sampling is considered time consuming and laborious in terms of drawing a sample. Incorrect Feedback: Rationale (INCORRECT): 1. Convenience sampling is considered very easy in terms of drawing a sample. Rationale (INCORRECT): 2. Quota sampling is considered relatively easy in terms of drawing a sample.

Which reliability measures are considered to be tests of stability? (Select all that apply). Item-to-total correlation Split-half reliability Interrater reliability Parallel, or alternate-form, reliability Test-retest reliability Kuder-Richardson coefficient Cronbach's alpha

Rationale (CORRECT): 4. Parallel, or alternate-form, reliability is considered a test of both stability and equivalence. Rationale (CORRECT): 5. Test-retest reliability is considered a test of stability.

Convergent validity

Refers to a search for other measures of the construct Established by exploring the correlations between the KPCS and selected subscales of the four other measures. The KPCS scores were associated (correlated) in the appropriate, demonstrating sound convergent validity.

Ordinal

Relative rankings Slightly better than nominal but still limited Frequencies, percentages, medians, percentiles, and rank-order coefficients can be calculated Ex: social class: high, middle, working

Which of the following can jeopardize the representativeness of the sample regardless of the technique used? (Select all that apply.) A large sample size Equal representation of each sample subset Sample heterogeneity Subjects dropping out of the study

Sample heterogeneity Subjects dropping out of the study

Bias is introduced into a study when: Sample selection is carried out inconsistently Subjects do not volunteer to participate Probability sampling techniques are used The population is homogeneous

Sample selection is carried out inconsistently Rationale (CORRECT): 1. Inconsistent sample selection jeopardizes the representativeness of the sample and introduces bias.

The process of selecting representative units of a population for a study in a research investigation

Sampling

Sampling threats to internal validity/external validity

Selection effect: -bias b/c of subject self-selection or research bias in selecting Attrition and mortality: -Loss of subjects from the study -If the subjects who remain in the study are different from those who drop out, the results can be affected External validity: generalizability to other populations or settings.

Using random numbers to select a sample from a list of possible subjects is called: Simple random sampling Stratified random sampling Nonprobability sampling Cluster sampling

Simple random sampling Rationale (CORRECT): 1. One technique of randomly selecting a sample is to use random numbers and select from a list of all possible subjects. Rationale (INCORRECT): 2. Stratified random sampling involves dividing the target population into strata or subgroups. Rationale (INCORRECT): 3. Random sampling is not a type of nonprobability sampling. Rationale (INCORRECT): 4. Cluster sampling is a multistage process.

An operational definition: Specifies how a concept will be measured Conveys the general meaning of a concept Is the most abstract definition of a concept Is a dictionary definition

Specifies how a concept will be measured Rationale (CORRECT): 1. An operational definition specifies how the concept will be measured and what instruments will be used to capture the concept.

Structured observational methods

Specify in advance what behaviors or events are to be observed Standardized forms are used for record-keeping and include categorization systems, checklists, or rating scales Rely heavily on the formal training and standardization of the observers

A researcher divides a questionnaire in half, with the even-numbered items in one group and the odd-numbered items in a second group. A reliability coefficient is calculated for the two halves. What method of determining reliability is the researcher using? Item-to-total correlations Split-half reliability Cronbach's alpha KR-20 coefficient

Split-half reliability Rationale (CORRECT): 2. Split-half reliability involves dividing a scale into two halves and making a comparison.

Which sampling strategy contains enhanced representativeness? Convenience Quota Simple random Stratified random

Stratified random Rationale (CORRECT): 4. Stratified random sampling is considered to have enhanced representativeness of the sample. Rationale (INCORRECT): 1. Convenience sampling is not considered to have enhanced representativeness of the sample. Rationale (INCORRECT): 2. Quota sampling does not have enhanced representativeness of the sample. Rationale (INCORRECT): 3. Simple random sampling does not have enhanced representativeness of the sample.

What are potential disadvantages of longitudinal studies? Subject loss as a result of attrition Loss of a control group The Hawthorne effect Superficiality of data

Subject loss as a result of attrition The Hawthorne effect

In the classic or true experimental design: Subjects are randomly assigned to one of two groups, each group receives treatment or intervention, and then the dependent variable is measured. Subjects are randomly assigned to one of two groups, each group receives a pretest and then treatment or intervention, and each group receives a posttest. Subjects are randomly assigned to one of two groups, the dependent variable is measured in both groups, one group receives treatment, and both groups are posttested. Subjects are randomly assigned to one of two groups, both groups are given a pretest, one group receives treatment, and the dependent variable is measured in one group.

Subjects are randomly assigned to one of two groups, the dependent variable is measured in both groups, one group receives treatment, and both groups are posttested. Rationale (CORRECT): 3. In an experimental design, subjects are randomly assigned to one of two groups to control for antecedent variables. Both groups are given a pretest. One group receives treatment or intervention. Both groups are posttested.

Normal distribution

Symmetric distribution Bell-shaped About 68% of the scores of means will fall within 1 standard deviation (SD) 95% within 2 SD of the mean 99.7% within 3 SD of the mean Scores more concentrated in the middle than in the tails Unimodal

What is a normal curve? (Select all that apply.) Symmetrical about the mean Unimodal Multimodal Skewed to the right

Symmetrical about the mean Unimodal

The statistical result of a research study is reported as "t = 2.65, p = 0.01." Which statistical test produces these results? t test Analysis of variance Chi-square Pearson correlation

T-test

Major tests of reliability:

Test-retest Parallel or alternate form Item-total correlation Split-half Kuder-Richardson (KR-20) Cronbach's alpha Interrater reliability

Which reliability measure is considered to be a test of stability? Split-half reliability Test-retest reliability Parallel, or alternate-form, reliability Cronbach's alpha

Test-retest reliability Rationale (CORRECT): 2. Test-retest reliability is considered a test of stability. INCORRECT: Split-half reliability is considered a test of homogeneity. Rationale (INCORRECT): 3. Parallel, or alternate-form, reliability is considered a test of equivalence. Rationale (INCORRECT): 4. Cronbach's alpha is considered a test of homogeneity.

A Cronbach's alpha coefficient of 0.80 indicates: That the frequency in each category is different from what would be expected by chance That the scale actually measures the concepts it was intended to measure That the instrument used is homogeneous That the data are collected consistently

That the instrument used is homogeneous Rationale (CORRECT): 3. Cronbach's alpha measures the homogeneity of an instrument with a Likert-type format. Rationale (INCORRECT): 1. Chi-square measurement is used to determine whether the frequency in each category is different from what would be expected by chance. Rationale (INCORRECT): 2. Factor analysis is used to determine whether a scale actually measures the concepts that it is intended to measure. Rationale (INCORRECT): 4. Consistency among observers using the same instrument tool is called equivalence and is not measured by Cronbach's alpha.

What is internal validity? The use of expert opinion to judge the accuracy of an instrument The degree to which findings of a study can be generalized to other populations The degree to which it can be inferred that the experimental treatment resulted in the observed effects The degree to which an instrument is said to measure a theoretical construct or trait

The degree to which it can be inferred that the experimental treatment resulted in the observed effects Rationale (CORRECT): 3. Internal validity is the degree to which it can be inferred that the experimental treatment resulted in the observed effects.

Which of the following should be considered when estimating a sample size? (Select all that apply.) The design of the study The time available for analyzing data The type of sampling procedure used The cost of collecting data

The design of the study The time available for analyzing data The type of sampling procedure used The cost of collecting data

Which of the following statements are true about convenience sampling? (Select all that apply.) Convenience sampling is the most commonly used method of sampling. Convenience sampling is the strongest nonprobability sampling strategy. Research consumers should be skeptical about the external validity of the finding. Convenience sampling provides level III evidence.

The easiest and most commonly used method of sampling is convenience sampling. Rationale (CORRECT): 3. Convenience sampling is the weakest form of sampling strategy with regard to strength of evidence and generalizability and is at highest risk for problems with external validity. PROVIDES LEVEL VI EVIDENCE

Target population

The entire set of cases about which the researcher would like to make generalizations Ex: might include all cholecystectomies, performed on female patients in the United States

Construct validity

The extent to which a test measures a theoretical construct or trait and attempts to validate a body of theory underlying the measurement and hypothesizing of relationships Subtypes: hypothesis-testing approach, convergent, divergent, and multi-trait-multimethod approaches, contrasted group approach, factor analytical approach.

In one retrospective study, a researcher demonstrated that cigarette smoking appeared to be a determinant of lung cancer. Why was the researcher unable to conclude a causal relationship between cigarette smoking and lung cancer? (Select all that apply.) The independent variable was not manipulated. Subjects were not randomly assigned to groups. The researcher did not have a high level of control. The causal event did not occur before the effect event

The independent variable was not manipulated. Subjects were not randomly assigned to groups.

There are advantages to using directional hypotheses, including (select all that apply): The indication of the use of a theory base to derive the hypothesis The provision of a specific theoretical frame of reference Ensuring that findings will be generalizable The indication of a nonbiased selection of subjects

The indication of the use of a theory base to derive the hypothesis CorrectThe provision of a specific theoretical frame of reference

The precision and sensitivity of physiological measurements are compromised when: Only one investigator uses a measuring device to collect data Subjects are familiar with the measuring device The measuring device is affected by the environment Equipment is too expensive to obtain and use

The measuring device is affected by the environment

An element is

The most basic unit about which information is collected In nursing research, elements are usually individuals A sample is a set of elements that make up the population

When differences between scores are too large to be caused by chance: The null hypothesis is rejected. The scientific hypothesis is proven to be true. The intervention did not make a difference. The scientific hypothesis is rejected.

The null hypothesis is rejected. Rationale (CORRECT): 1. The null hypothesis (the hypothesis that there are no differences) is rejected when there are differences between variables.

How are predictive models used? To synthesize findings from randomized clinical trials to test the effectiveness of an intervention To develop measurement techniques to establish causal relationships To generalize findings from nonexperimental studies To explain the relationships among variables to establish predictive or causal link

To explain the relationships among variables to establish predictive or causal links Rationale (CORRECT): 4. Predictive models are used to determine predictive relationships.

When data are positively skewed, what is the mean? To the right of the median To the left of the median The same as the mode Closer to zero than the mode

To the right of the median Rationale (CORRECT): 1. With skew, the data are distributed asymmetrically and have the peak off center. A positive skew means that the tail of the curve is to the left and the mean is to the right of the middle (median) score.

How are psychometrics used? To interpret research data to determine cause-and-effect relationships To use the research process to develop tools or techniques for use in measurement To combine data from multiple research studies and summarize their findings To analyze data collected for another purpose

To use the research process to develop tools or techniques for use in measurement Rationale (CORRECT): 2. Psychometrics deals with the theory and development of measurement and instrumentation.

Criterion-related validity

To what degree the subject's performance on a measurement tool and the subject's actual behavior are related. Subtypes: concurrent validity and predictive

T/F Experimental and Quasi-experimental designs use manipulation

True

T/F internal validity must be established before external validity

True!

Z-score

Used to compare measurements in standard units Each of the scores is converted to a Z-score, and then the Z scores are used to examine the relative distance of the scores from the mean

Existing Data

Uses records and databases For example, medical records, care plans, hospital records, death certificates, U.S. Census, National Cancer Data Base Often referred to as secondary analysis

Which critical appraisal questions would be applicable to data analysis? (Select all that apply). How have the rights of subjects been protected? Were the descriptive or inferential statistics appropriate to the level of measurement for each variable? Did testing of the hypothesis(es) or research question(s) clearly support or not support each hypothesis or research question? What are the limitations in generalizability? Is a rationale given for why a particular instrument or method was selected? To what population may the findings be generalized? Are the inferential statistics appropriate for the type of design, hypothesis(es) or research question(s)? What provision is made for maintaining accuracy of the instrument and its use, if any?

Were the descriptive or inferential statistics appropriate to the level of measurement for each variable? Did testing of the hypothesis(es) or research question(s) clearly support or not support each hypothesis or research question? Are the inferential statistics appropriate for the type of design, hypothesis(es) or research question(s)?

Which critiquing criterion would apply to observational methods? Were the observers trained to minimize any bias? Are operational definitions provided and clear? Is there clear indication that the subjects understood the task and the questions? Is there evidence that subjects were able to answer the questions?

Were the observers trained to minimize any bias? Rationale (CORRECT): 1. This critiquing criterion would apply to observational methods.

Which critical appraisal questions would be applicable to sampling? (Select all that apply). How have the rights of subjects been protected? Were the descriptive or inferential statistics appropriate to the level of measurement for each variable? Did testing of the hypothesis(es) or research question(s) clearly support or not support each hypothesis or research question? What are the limitations in generalizability? Is a rationale given for why a particular instrument or method was selected? To what population may the findings be generalized? Are the inferential statistics appropriate for the type of design, hypothesis(es) or research question(s)? What provision is made for maintaining accuracy of the instrument and its use, if any?

What are the limitations in generalizability? To what population may the findings be generalized?

Which critical appraisal question would be categorized as pertaining to internal and external validity? Given the level of evidence suggested by the research question, hypothesis, and design, what is the potential applicability to practice? Is there a synthesis summary that presents the overall strengths and weaknesses and arrives at a logical conclusion that generates hypotheses or research questions? Would it be possible to replicate this study in another clinical setting? What are the sources of bias and are they dealt with appropriately?

What are the sources of bias and are they dealt with appropriately? Rationale (CORRECT): 4.This would most accurately categorize internal and external validity.

When is purposive sampling useful? When few people are willing to participate When members of the target population are heterogeneous When a highly unusual group is being studied When many subjects drop out of a study before it is completed

When a highly unusual group is being studied Rationale (CORRECT): 3. When few people meet the selection criteria, researchers often form a sample from all those eligible people who are willing to participate.

When are the precision and sensitivity of physiologic measurements compromised? When only one investigator uses a measuring device to collect data When subjects are familiar with the measuring device When the measuring device is affected by the environment When equipment is too expensive to obtain and use

When the measuring device is affected by the environment

Which critical appraisal question would be categorized as applicability to nursing practice? Given the level of evidence suggested by the research question, hypothesis, and design, what is the potential applicability to practice? Is there a synthesis summary that presents the overall strengths and weaknesses and arrives at a logical conclusion that generates hypotheses or research questions? Would it be possible to replicate this study in another clinical setting? What are the sources of bias and are they dealt with appropriately

Would it be possible to replicate this study in another clinical setting? Rationale (CORRECT): 3. This would most accurately categorize applicability to nursing practice.

Non-equivalent control group design (Quasi-experimental design)

-Experimental and control group, but NO randomization -Threats to internal validity include: selection, maturation, testing, and mortality -Results of the study still relatively strong because the researcher is able to compare the two groups on antecendent variables before the intervention.

Means of control

-Homogenous sample -Consistent data-collection procedures -Manipulation of IV -Randomization

Disadvantages of surveys

-Info obtained tends to be broad but superficial -Conducting a survey requires expertise in several research areas -Large-scale surveys can be time-consuming and costly

Control

-Introducing one or more constants into the experimental situation -Control is acquired by: Manipulating the IV, Randomly assignment, using a control group -The control group receives the usual treatment or placebo

Advantages of Cross-Sectional studies

-Less time consuming and expensive -Large amounts of data can be collected at one point, making the results more readily available -The cofounding variable of maturation, resulting from elapsed time is not present

What is control?

-Measures that the researcher uses to hold the conditions of the study uniform and avoid possible impingement of bias on the dependent variable or outcome variable. -Uniformity over tested variables

Strengths of experimental design

-Most powerful for testing cause-and-effect relationships owing to the use of control, manipulation, and randomization

Strengths of Quasi-Experimental Designs

-Practical, less expensive, generalizable, and sometimes the only feasible alternative because these designs are more adaptable to real-world settings -Replication of a study can strengthen evidence

Level II evidence

A well-designed RCT

A researcher studies the differences in IQ between girls and boys. She tests cohorts of girls and boys in the third grade and then tests the same girls and boys when they are in the sixth grade. Which of the following are threats to internal validity in this study? (Select all that apply.) A. Maturation B. History C. Testing D. Measurement effects

A. Maturation C. Testing

In quasiexperimental designs: The threats to external validity are numerous. The level of evidence obtained is II. One of the characteristics of a true experiment is lacking. The findings are not as valuable as those of experimental designs.

One of the characteristics of a true experiment is lacking. Rationale Quasiexperimental designs lack manipulation, control, or randomization. Quasiexperimental designs have more threats to internal validity. Quasiexperimental designs yield Level III evidence. The value of a quasiexperimental study is not lessened, although the strength of the findings may be less.

After-only Nonequivalent control group design (Quasi-experimental design)

-Randomization is not used and makes the assumption that the experimental and control groups are equivalent before the introduction of the IV -Confidence in the findings depends on the integrity of this assumption that the two groups are comparable.

Experimental and Quasi-Experimental designs

-Require that the researcher actively intervenes to bring about the desired effect -Test cause-and-effect relationships -Provide level II and level III evidence

Accuracy in the research design means that: A. All aspects of the study logically flow from the research question. B. All mediating variables have been controlled. C. Gaps in the literature have been identified. D. The timing of data collection was the same for each subject.

A. All aspects of the study logically flow from the research question.

All aspects of a study systematically and logically flow from a research question

Accuracy

In quantitative research, the purpose of control is to avoid: A. Duplication. B. Bias. C. Randomization. D. Homogeneity.

B. Bias.

Surveys

Collect detailed descriptions of current or existing variables and use the data to justify and assess current conditions and practices or to improve health care practices. Provide the basis for program development and interventions Samples may be small, large, broad or narrow Surveys do not attempt to determine causation* Seek to relate variables to each other

How does randomization eliminate systematic bias in groups of subjects? The independent variables are manipulated by the researcher. Each subject has an equal chance of being changed by the dependent variable. Extraneous variables are equally distributed among groups. Each subject is less likely to be impacted by extraneous variables.

Extraneous variables are equally distributed among groups. Rationale (CORRECT): 3. By randomly selecting subjects, there is a chance that extraneous variables are equally distributed among groups and will have an equal effect on the dependent variable.

What level of evidence do experimental studies provide? II IV VI VII

II

When data are collected multiple times before and after the introduction of the treatment, the study: (Select all that apply.) Is called a time series design. Cannot be called a quasiexperimental study. Is a nonequivalent control group design. May have threats of selection and maturation.

Is called a time series design. May have threats of selection and maturation.

A true experimental design has which of the following properties? (Select all that apply.) Manipulation Relevance Randomization Control

Manipulation, Randomization, and control

In the Solomon four-group design: All groups take the pretest. All groups are considered experimental groups. Two groups take the pretest and the other two groups take the posttest. Only two groups take the pretest

Only two groups take the pretest Rationale (CORRECT): 4. Of the four groups, two are control groups and two are experimental groups.

In the classic or true experimental design: Subjects are randomly assigned to one of two groups, each group receives treatment or intervention, and then the dependent variable is measured. Subjects are randomly assigned to one of two groups, each group receives a pretest, each group receives treatment or intervention, and each group receives a posttest. Subjects are randomly assigned to one of two groups, the dependent variable is measured in both groups, one group receives treatment, and both groups are posttested. Subjects are randomly assigned to one of two groups, both groups are given a pretest, one group receives treatment, and the dependent variable is measured in one group.

Subjects are randomly assigned to one of two groups, the dependent variable is measured in both groups, one group receives treatment, and both groups are posttested. Rationale (CORRECT): 3. In an experimental design, subjects are randomly assigned to one of two groups to control for antecedent variables. Both groups are given a pretest. One group receives a treatment or intervention. Both groups are posttested.

Solomon Four-Group Design

Subjects randomly assigned to 4 groups: experimental, control, experimental after group, and control after group -The experimental after group and control after group receive NO pretests, only posttests** -The design controls for TESTING effects that are a threat to internal validity **more expensive cause it requires a larger sample

Highest level of evidence: Level I

Systematic review or meta-analysis of RCT's

Controlling for extraneous variables and eliminating potential alternative explanations for study findings mean eliminating: Threats to validity. Independent variables. A priori control. Randomization of subjects.

Threats to validity.

T/F RCT is a strong design for testing cause-and-effect relationships?

True

Variables may be classified as

attitudes, opinions or facts

Meta-Analysis

-Meta analysis of RCT yields level I evidence -Uses statistical techniques to summarize and assess studies. -Offers stronger evidence because it is a rigorous process of evidence summarization. -Provide the most powerful and useful evidence available to guide practice--Level I

Purpose of Research design

-vehicle for testing research questions and hypotheses -involves structure and strategy to maintain control and intervention fidelity -aid in the systematic solution of research questions and hypotheses

Examples of extraneous variables include: (Select all that apply.) Antecedent variables. Cause and effect variables. Intervening variables. Manipulated variables.

Antecedent variables: affect the dependent variable before the introduction of the independent variable. Intervening variables: extraneous variables that occur during a study.

Level III evidence

Controlled trial without randomization (quasiexperimental study)

Control is defined as: A. Having a large sample size. B. Manipulating the dependent variable. C. Having a basis of comparison for each experimental group of subjects. D. Holding conditions of the study constant.

D. Holding conditions of the study constant.

The purpose of randomly selecting subjects is to: Increase the number of participants. Eliminate bias. Prevent contamination of subjects. Control the dependent variables.

Eliminate bias.

Manipulation of IV is done by

Experimental groups and control groups

Which type of design is most powerful for providing information about cause-and-effect relationships? Randomized clinical trial Quasiexperimental Nonequivalent control group Time series

Randomized clinical trial

A nurse, conducting a clinical trial, administers a pain medication to an experimental group and a placebo (sugar pill) to the control group. To her surprise, both groups experienced a significant reduction in pain. This may be due to: Maturation Reactivity Measurement effects. Instrumentation.

Reactivity. Rationale (CORRECT): 2. Reactivity refers to the subject's response to being studied. In this case, the Hawthorne effect changed the dependent variable.

Advantages of Longitudional-Prospective Cohort studies-Developmental study #2

-*Each subject is followed up seperately and thereby serves as his or her own control -Increased depth of responses can be obtained -Early trends in the data can be analyzed -Assesses changes in the variables over time, and both relationships and differences can be explored between variables.

Longitudional-Prospective Cohort studies-Developmental study #2

-Collect data from the same group at DIFFERENT points in time -Begin in the present and end in the future -Also referred to as repeated measures or cohort studies -Allows for assessment of the incidence of a problem over time and potential reasons for changes in the variables of study. -Ex: To examine the pain experience management strategies and outcomes during the first year after diagnosis of acute leukemia.

Integrative reviews

-Critically appraise the literature in an area without statistical analysis. -Broadest category of review -Can include either or both theoretical and research literature.

Developmental studies (3) Retrospective-Ex Post Facto-Case control studies

-DV has already been affected by the IV. -Investigator attempts to link present events to events that occurred in the past: linking smoking to lung cancer is a good example of a retrospective. -Also known as case control, causal-comparative, or comparative studies -Higher level of control than a correlational study -Linking smoking to lung cancer is a good example of a retrospective study. -Hypothesis: smoking (independent variable-x) is related to and a determinant of lung cancer (dependent variable-Y) -Subjects in the control group resemble the smoking group in every way except one; they do not smoke -The behavior, performance, or condition (lung tissue) of the two groups is compared to determine whether the exposure to X had the effect (lung cancer) predicted by the hypothesis.

Disadvantages of Longitudional-Prospective Cohort studies-Developmental study #2

-Data collection may take a long time, increasing costs in time, effort, and money -Testing effects may be a threat -Mortality is a significant threat owing to the increased potential for attrition.

Methodologic studies

-Development and evaluation of data-collection instruments, scales, or techniques -Psychometrics is the theory and development of measurement instruments (such as questionnaires) or measurement techniques (such as observational techniques) -Psychometrics measures concepts like anxiety or quality of life

Disadvantages of Cross-Sectional studies

-Difficult to establish an in-depth developmental assessment of the interrelationships of the phenomena

Disadvantages of Retrospective-Ex Post Facto-Case control studies

-Difficulty explaining cause-and-effect relationships -Important to developing knowledge base about phenomenon of interest. -Useful in forecasting or making predictions

Correlational studies

-Do not test cause-and-effect -Examine relationship between 2 or more variables -Testing whether variables co-vary-as one variable changes, does a related change occur in the other variable? -Quantifying the strength and the direction (positive or negative) of a relationship between variables or in testing a hypothesis about a specific relationship. -A framework for exploring relationships between variables that cannot be inherently manipulated

Randomization

-Each subject in the study has an equal chance of being assigned to the control group or the experimental group -Assumes that any important intervening variable will be equally distributed between the groups minimizing variance and decreasing selection bias -An intervening variable is a variable that occurs during the study that affects the dependent variable

Cross-Sectional Studies (developmental study #1)

-Look at a broader perspective of a cross-section of the population at a specific time with the same subjects -Examine data at one point in time** -Can be one group or simultaneously collecting data on the study's variables from different groups. -Explore relationships, correlations, differences or comparisons. Ex: to examine the influence of stroke survivors' motor function, memory, and behavior changes as well as family conflict surrounding stroke recovery on the mental and physical health of caregivers during the subacute recovery period.

Non-experimental designs

-Provides level IV evidence -Used in studies where the researcher wishes to construct a picture of a phenomenon, explore events, people, or situations as they naturally occur; tests relationships among variables -May study at one point or over a period of time -IV is NOT manipulated -IVs have naturally occurred; investigator cannot directly control them by manipulation -Requires a clear, concise research problem or hypothesis based on theoretical framework -Subjects are NOT randomized -There is no control group

External validity

-Questions under what conditions and what types of subjects the same results can be expected to occur. -Questions the conditions under which the findings can be generalized to other populations/environments

Summary of the quantitative research literature that used similar designs based on a focused clinical questions

-Systematic reviews

Randomized clinical trial (RCT)

-Use experimental and control groups -Have a very specific sampling plan, using inclusion and exclusion criteria -Intervention fidelity ensures that every subject receiving the intervention receives the identical intervention -Uses statistical comparisons to determine any differences between groups

One group Pretest-Posttest design (Quasi-experimental design)

-Used by researchers when only one group is available for study -Data are collected before and after an experimental treatment on one group of subjects **NO control group/no randomization

Time Series Design (Quasi-experimental design)

-Used when only ONE group is available to study over a longer period of time -Useful for determining trends over time -Data are collected multiple times before the introduction of the treatment ot establish a baseline point of reference on outcomes. -The experimental treatment is introduced and data are collected multiple times afterward to determine a change from baseline. -The broad range and number of data-collection points helps rule out alternative explanations, such as history effects. -Potential threats to Internal validity: Testing--b/c of multiple data-collection points, maturation and selection--b/c lack of control group

Systematic reviews

-no statistical techniques used to summarize and assess studies. -review of quantitative research that used similar designs and focuses on a particular clinical question. -Systematic review or meta-analysis of RCTS=Level I evidence -Systematic reviews of descriptive and qualitative studies yields level V evidence. -Uses rigorous inclusion/exclusion criteria, assesses sampling issues, threats to internal validity (bias), external validity, and the results. -Report current and valid research on intervention effectiveness and clinical knowledge in order to develop EBP -If statistical methods are used to analyze the findings, the systematic review is then called a META-ANALYSIS

A study has internal validity if:

-the IV (intervention) is the source of a difference or change in the DV (variable of interest)

3 types of Survey studies (non-experimental)

1. Descriptive-describe the characteristics of particular subjects, groups, institutions, or situations or frequency of a phenomenon's occurrence. 2. Exploratory-Explore the characteristics of particular subjects, groups, institutions or situations 3. Comparative-Determine differences between variables. Collects detailed descriptions of existing variables. Data collected through questionnaires or interviews.

4 types of Quasi-experimental design

1. Nonequivalent control group design 2. (After-only) Nonequivalent control group design 3. One-group (Pretest-Postest) design 4. Time series design

3 Properties of true experimental design:

1. Randomization of subjects 2. Control of extraneous variables 3. Manipulation of IV

Classic RCT's usually follow these 4 steps:

1. Representative sample is recruited from the population of interest 2. Demographics and measurement of the variables of interest (dependent variables) are collected before the intervention 3. Participants are randomized to the experimental or control group

Which of the following are means of controlling extraneous variables? (Select all that apply.) A. Randomly selecting subjects B. Using a homogeneous sample C. Manipulating the dependent variable D. Conducting a pilot test

A. Randomly selecting subjects B. Using a homogeneous sample

Administration of an intervention to one group of subjects and not another is an example of: (Select all that apply.) A. Homogeneity of subjects. B. Manipulation of the independent variable. C. An experimental study. D. The introduction of bias.

B. Manipulation of the independent variable. C. An experimental study.

For large experimental studies, conducting a pilot study helps the researcher determine the: A. Degree of internal control over dependent variables. B. Accuracy and objectivity of the design. C. Feasibility of the research design. D. Constancy of data collection.

C. Feasibility of the research design.

In order to test cause-and-effect relationships, the: Causal variable and the effect variable cannot be associated with each other. Cause must precede the effect. Cause-effect relationship must be associated with the extraneous variables. Effect variable must be an independent variable.

Cause must precede the effect.

T/F Non-experimental designs manipulate the IV

False--Non-experimental designs do NOT manipulate the IV

Fact variables may include

Gender, income level, political and religious affiliations, ethnicity, occupation, and education level

Studies conducted in laboratories have: Higher internal validity and lower external validity than field experiments. Higher internal validity and higher external validity than field experiments. Lower internal validity and lower external validity than field experiments. Lower internal validity and higher external validity than field experiments.

Higher internal validity and lower external validity than field experiments. Rationale (CORRECT): 1. The research has almost total control over the research environment, giving the study more internal validity, but less control over factors specific to the setting in which the findings will be applied, giving the study lower external validity.

A research study was underway looking at the frequency of mammograms in perimenopausal women. During the study a celebrity was diagnosed with breast cancer. The effect of this event on the research findings is called: Maturation. Reactivity. Constancy. History.

History--This event occurred outside the experimental setting and may have had an effect on the dependent variable.

Ensures that every subject receiving the intervention or treatment receive the identical intervention or treatment

Intervention fidelity

Threats to external validity

MRS. M-Measurement effects--administration of a pre-test affects generalizability of the findings to other populations. A pre-test may act like an intervention and influence or change the subjects' responses. R-Reactive effects-Subjects response to being studied. Subjects may respond to the investigator not b/c of the study procedures, but merely as an independent response to being studied. S-Selection effects--What method was used to select the sample? How were subjects assigned to groups?

Which of the following are means by which control is introduced into the research design? (Select all that apply.) Manipulation of the dependent variable Manipulation of the independent variable Randomization Use of control groups

Manipulation of the independent variable Randomization Use of control groups

A researcher studies a group of freshman college students to measure the impact of an HIV/AIDS awareness program on their understanding of the disease. Students attended the program and then were given a 10-question test about HIV/AIDS. Which of the following could be an antecedent variable threatening the internal validity of this study? A televised, public education program started during this period of time. Students had varying degrees of knowledge about HIV/AIDS before participating in the study. Several of the subjects were actually diagnosed with HIV/AIDS during the study. Twenty percent of the subjects dropped out of school before taking the test.

Students had varying degrees of knowledge about HIV/AIDS before participating in the study. Rationale (CORRECT): 2. Prior knowledge of the students is an antecedent variable and could threaten the internal validity of this study.

Threats to internal validity

THIMMS T-Testing H-History I-Instrumentation M-Maturation M-Mortality S-Selection bias

When two groups, control and experimental, take a posttest after the intervention without taking a pretest: There is a threat to external validity. The design is called the after-only design. Mortality is lower. There are fewer extraneous variables.

The design is called the after-only design.

Internal validity asks: If the study is generalizable to the larger population. If the Hawthorne effect affected the findings. Whether the independent variable really made the change in the dependent variable. Whether study findings are a result of measurement effects.

Whether the independent variable really made the change in the dependent variable.