inclusion criteria

exclusion criteria

process of selecting a part of the population to represent the entire population for a particular study

narrow the population to a researchable size and characteristics that define the population

the subset of the population that is chosen to represent the population for a given study

involves random selection of elements: each element has an equal, independent chance of being selected

elements chosen in nonrandom way; means not every element of population has opportunity to be chosen for the sample

accidental/incidental; subjects for study are chosen as they become available

weakest form of sampling

a sample whose key characteristics closely approximate those of the population

researcher takes into account some knowledge of population to build in representativeness to the sampling plan

researcher wants to study patient satisfaction with emergency department nursing care

the systematic overrepresentation or underrepresentation of some segment of the population in terms of key characteristics

recruit all of the people from those that are accessible who meet the eligibility criteria, (1) until the specified sample is obtained or (2) the specific time interval is completed

researcher "hand picks" the subjects for the study; most often used in qualitative research

researcher only selects persons who have experienced loss due to a Hurricane

researcher uses subjects to access other potential subjects; used to identify people with distinctive characteristics

a researcher wishes to study health habits of elderly people living in a retirement community. the researcher has one subject and then asks that subject to refer other people in the building to the study

which type of sampling is most vulnerable to bias

does not involve selection of elements at random

subjects selected in random fashion; each element of the population has an equal independent chance of being chosen for the study

researcher obtains list of all elements in the population (sampling frame); assigns a number to each element

size is in proportion to their representation in the population

same as non-probability quota sampling except after homogeneous subsets or strata are established, elements for each stratum are chosen by simple random sampling

different proportions of study participants sampled when comparing strata; helps make sure smaller strata are adequately represented

involves successive random sampling of units that progresses from large groups or clusters to smaller groups or clusters

researcher wants to generalize to all Baccalaureate nursing student in US

the researcher chooses each element of the population from a list or group

researcher wants to sample all emergency department patients seen in last year

all insulin dependent diabetics in Cabell County

stratified random sampling enhances representativeness

as sample size increases, sampling error

determines how large the sample size must be to detect a difference

male/female, chronically ill/acutely ill

risk of getting it wrong increases if sample is too small

method of processing and analyzing numbers in systematic manner

subpopulations of a population

used when research problem does not lend itself to an experimental design

those within the subjects; characteristics of the study participants

age, gender, underlying medical illness, personality traits, emotions, perceptions of an event, etc

those within the subjects; characteristics of the study participants

sex, blood type, age, etc

randomly selecting subjects for each group; each member has an equal chance of being selected

smoking

sample too small

independent variable not powerful

most effective means of controlling internal confounding variables; confounding variables will be distributed over both groups evenly

unreliable implementation of a treatment

examine internal and external validity to evaluate the adequacy of the research design

when effects detected in a study are a true reflection of reality and not the result of the effects of extraneous variables

when some event other than the experimental intervention which might influence the dependent variable; ex. news program on breast self-examination

occurs when groups differ before intervention in characteristics; ex. smoking

if it is impractical to manipulate independent variables such as lack of time or lack of funds

occurs when natural changes occur within subjects during a study which might affect results; ex. school nurse study of nutritional status of children

the function of random assignment is to choose a comparison group that is as much like the intervention group as possible

controlled by use of comparison group and random assignment

occurs when pretest is given and the subjects may remember the answers

occurs when different pretest and posttest instruments are used; best if the instruments have similar quality assessment values

occurs when subject dropout rate is different between intervention and comparison groups; difficult to control

concerns to what extent to which the findings of a study can be generalized (applied to other people or settings)

concerns whether the sample is representative of population is an inadequacy of sampling design threat

ascertaining the prevalence of a health problem or trait

occurs when subjects in a study change behavior simply because they know they are in a study

used when random assignment not feasible; researcher chooses for the study only those subjects who are alike with respect to the extraneous variables

occurs when characteristics or behavior of the researcher affect behavior or responses of subjects

use subject characteristics to match subjects on one-on-one basis to make groups comparable on extraneous variables such as age, gender, and race

clothes, facial expressions, gender, and body build

occurs when subjects are pre-tested; researchers cannot generalize people who were not pre-tested

data must not be influenced by data collection

the purpose is to describe whether variables are related, without ascribing a cause-and-effect connection

everyone must collect data in the same way

existing data

data which the researcher prospectively collects

randomly select subject in the experimental and then a subject that is similar to that subject is randomly selected for the control group

historical data that is convenient and economical; secondary or meta-analysis

assessment of clinical measures (e.g. blood pressure, O2 saturation, potassium level)

purpose is to observe, describe, and explain aspects of situation

subject's perceptions (e.g. interview)

what size pool of subjects do you need for matching

direct questioning of study subjects

can control for confounding variables using various statistical methods

researcher's perceptions (e.g. direct observation of behavior)

the ability to detect true relationships statistically

use of special equipment to make measurements; complex or simple equipment

may compare two or more groups on an attribute but no intervention

performed directly within or on living organisms (e.g. BP)

the extent to which it can be inferred that the independent variable caused or influenced the deoendent variable

biophysiologic material is taken and analyzed outside the body (e.g. blood chemistry levels, cytology, etc)

cause-probing questions for which manipulation is not possible are typically addressed with

data are collected with a formal instrument

the generalizability of the observed relationships across samples, settings, or time

quantitative structured self-reports

allows persons to respond in their own words in a narrative fashion

an association between variables and can be detected through statistical analysis

provide answers to choose from

when items are combined to obtain an overall score

used to make fine quantitative discriminations among people with different attitudes, perceptions, and traits

cause-probing questions (e.g. prognosis or harm/etiology questions) for which manipulation is not possible are typically addressed with a correlational design

summated rating scales; respondents indicate their view on a topic

consists of several declarative statements; responses are on an agree/disagree continuum

assign numbers to represent the amount of an attribute of characteristic present in a person or object

obtained if no error in measurement; never known

measure obtained

amount of random error in the measurement process

causes a person's observed score to vary in no particular manner around their true score

rate concepts using bipolar adjectives

random error does not influence direction of the mean but increased the amount of unexplained variance

not random in nature and occurs due to measurement of something else in addition to concept

uses a continuum of bipolar adjectives (good/bad, strong/weak), and respondents are asked to rate a given concept, situation, or experience on a scale between each set of adjectives

scale not properly calibrated therefore subject's weights were lower or higher than true weight

systematic error can affect the mean score

on a descriptive continuum, typically bipolar

effects of the environmental factors on scores; temperature, lighting, time of day

includes characteristics of the subjects that interfere with accurate measurement; socially desirable responses, extreme responses, education

give answers consistent with prevailing views in society

factors with the subjects which can affect measurement; fatigue, hunger, anxiety, mood

refers to whether the instrument looks as though it is an appropriate measure of the construct; based on judgement

determines that the scale adequately examines the domain of the concept, all items are relevant, and all aspects are included

indicator of the degree of instrument's validity based upon the average of ratings by the expert panel

consistent expression using extreme responses (e.g. strongly agree)

broad term which encompasses other types of validity and determines if the instrument measures the construct it says it measures

statistical method for testing construct validity; closely related items cluster together into factors

calculated by analyzing the relationship between scores on the instrument and the criterion; desirable is 0.70 or higher

occur when variations in intervention administration or procedure for collecting data from one subject to another affect measurement; arm not in same position for BP, change in test questions by interviewer

the instrument's ability to distinguish people whose performance differs on a future criterion

the instrument's ability to distinguish individuals who differ on a present criterion

the difference between predictive validity and concurrent validity is

measure subjective experiences such as pain, nausea, vomiting, grief, and anxiety

differentiate between score on GRE and grades in graduate school

score on the Beck Depression Inventory and nurses ratings of severity of the patients symptoms of depression

occurs if instruments used for measuring are vague or poorly understood; different interpretations by various respondents due to questions not being clear

the instrument's ability to correctly identify a "case" - i.e. to diagnose a condition

result of poor sampling of items that are used to measure an attribute in instrument; score may vary as to what items are included on the test

involves a plan for choosing a group of people, events, behaviors or other elements for conducting a study

the instrument's ability to correctly identify noncases, that is, to screen out those without the condition

the process of making the selection

the true score is the score that would be obtained with an infallible measure

the selected group or elements chosen to represent to the entire population

complete set of persons, events, objects, that possess the characteristics the researcher is interested in studying

all insulin dependent diabetics

the portion of the target population that is accessible to the researcher, from which a sample is drawn

composed of entire group of people or objects in which researcher is interested

indicates how consistently and accurately the research instrument (such as scale) measures the variable of interest over time; ex. thermometer

exists in degrees and is expressed in the form of a coefficient ranging from 1.00 to 0.00

always in agreement with the statement regardless of content; nay-sayers always disagreement

1.00

refers to the consistency of a tool over time; similar scores each time it is used

0.00

a reliability coefficient less than 0.80 for a well developed instrument is

a reliability coefficient of 0.80 - 0.95 is preferred

evaluates stability over time; the same instrument is given to the same subjects at two points in time

stability reliability is not recommended for testing attributes that might

an outcome in the present (e.g. depression) is linked to a hypothesized cause occurring in the past (e.g. having had a miscarriage)

as one variable increases, the other variable increases (ex. increase in salt, increase in BP)

refers to whether the questionnaire is measuring what it says it is measuring

increase in one variable and decrease in other (ex. increase in height, decrease in BMI)

tests whether the instrument items measure the same attribute; requires only one test administration

all behaviors of a specific type recorded, and each behavior is assigned to one mutually exclusive category

requires only one test administration

assessed by coefficient alpha (Cronbach's alpha)

specific behaviors, but not all behaviors, recorded

0.00 to +1.00

"cases" (e.g. those with lung cancer) are compared to "controls" (e.g. those without lung cancer) on prior potential causes (e.g. smoking habits)

used to test agreement between/among different observers or interviewers

when determining the reliability of a measurement tool, which value would indicate that the tool is most reliable

reliability is lower in _____ than in _____ samples

formal systems for systematically recording the incidence or frequency of prespecified behaviors or events

reliability is lower in ____ than in ____ multi-item scales

the determination of the extent to which an instrument measures what it says it measures

a potential cause in the present (e.g. experiencing vs. not experiencing a miscarriage) is linked to a hypothesized later outcome (e.g. depression 6 months later)

validity is never proven, only supported

determine subjects behavior under different circumstances

stronger than retrospective in supporting causal inferences

sampling of time intervals for observation

neither prospective or retrospective designs are stronger than experimental designs

observation of integral events

observational methods for capturing many clinical phenomena and behaviors

factors that can interfere with objective observation; emotions, anticipation, personal views

looks to the past

potential problem of this when people are aware that they are being observed

looks to the future

same information in the same manner from all subjects

does not yield persuasive evidence for casual inferences

can be collected in a way to allow statistical analysis

strive to be objective not subjective

efficient way to collect large amounts of data when intervention and/or randomization is not possible

groups are formed through self selection therefore cannot assume the groups are equal or similar prior to the occurrence of the independent variable

data are collected at a single point in time

data are collected two or more times over an extended period

better at showing patterns of change and at clarifying whether a cause occurred before an effect (outcome)

challenge in longitudinal studies is _____ or the loss of participants over time

better at showing patterns of change and at clarifying whether a cause occurred before an effect (outcome) but are expensive, time-consuming, and have risk of attrition

plan for obtaining answers to research questions and/or for testing hypotheses

research design determines the amount of control the researcher exerts on the research situation

same people compared at different times or under different conditions

different people are compared (e.g. men and women)

when will information on independent and dependent variables be collected

retrospective

prospective

one way to control over confounding variables is to randomly assign subjects to either experimental or control group

most _____ research questions are about causes and effects

research questions that seek to illuminate causal relationships need to be addressed with appropriate designs

what would have happened to the same people exposed to a "cause" if they simultaneously were not exposed to the cause

represents the difference between what actually did happen when exposed to the cause and what would happen with the counterfactual condition

the cause must precede the effect in time

there must be a demonstrated association between the cause and the effect

the relationship between the presumed cause and effect cannot be explained by a third variable or confounder; another factor related to both the presumed cause and effect cannot be the "real" cause

the causal relationship should be consistent with evidence from basic physiologic studies

offer the strongest evidence of whether a cause (an intervention) results in an effect (a desired outcome)

true and quasi

descriptive and correlational

the researcher does something to some subjects - introduces an intervention (or treatment)

the researcher introduces controls, including the use of a control group counterfactual

the experimenter assigns participants to a control or experimental condition on a random basis

a true experimental design must have

which of the following would be a key criterion for causality

true experimental designs

allow for simultaneous manipulation of 2 or more variables at one time

researchers can examine the effect of combining the interventions with factorial designs

subjects are given more than one treatment sequentially rather than concurrently

subjects are exposed to 2+ conditions in random order

the effects of the treatment on the same subject are compared in crossover

whether the treatment as planned was actually delivered and received

control group gets no treatment at all

standard or normal procedures used to treat patients

example - auditory vs. visual stimulation

presumed to have no therapeutic value, but is used

extra attention, but not the active ingredient of the intervention

the intervention is given at a later date

most powerful method for testing cause and effect because researcher controls research situation

there is a high confidence in effectiveness of intervention with ____ experimental

the knowledge if being in a study may cause people to change their behavior

a true experiment requires that the researcher manipulate the independent variable by administering an experimental treatment (or intervention) to some subjects while withholding it from others

those getting the intervention are compared with a nonrandomized comparison group

involve an intervention but lack either random assignment or control group

one group is studied before and after the intervention and one group is studied after the intervention

one group studied over a period of time

involves 2 or more groups of subjects observed before and after the intervention

nonequivalent control group posttest only design is much weaker

two or more groups of subjects observed after the intervention

collect data over a period of time and introduce the IV or treatment during the data collection

one group of subjects observed before and after the intervention

typically yield extremely weak evidence of a causal relationship; lacks ways to control external factors that may result in differences

introduce the IV or treatment, then collect the data; one group of subjects observed after the intervention

involves one group of subjects observed over multiple times both before and after the intervention

do time series have a comparison group

to be classified as an experimental design (true or quasi) there MUST be

the design is automatically nonexperimental if there is

those that are a result of research situation

type of setting, time of day the data was collected, how the data is collected, etc