322 quiz 2

2) ordinal

2) SECOND LOWEST - a continuum of numeric values is used with small numbers representing lower levels on the continuum and larger numbers representing higher values - intervals are not meant to be equal ex) marathon runners who get 1st 2nd and 3rd do not have equal times and paces but there is value to their assigned number ex) on a scale from 1-5 how satisfied are you with ______ ex) letter grade

3) interaction of tx and selection of subjects 4) interaction of tx and setting

3) a threat to external validity where the independent variable might not affect individuals the same way- ex: condom use, target population is all sexually active teens- how generalizable are findings from an urban high school to rural high schools? 4) a threat to external validity when an intervention conducted in one setting cannot be generalized to a different setting ex: condom study conducted on teens in waiting room of family clinic generalizable to teens in high school population?

3) maturation 4) testing

3) during a LT study, subjects grow and get older-ex: looking at condom use and now the pt is 16 and can drive and get condoms 4) when a pretest influences the way subjects respond on a posttest- repeated testing can cause familiarity with the test itsself and answers may reflect subject's abilities to remember how questions were answered previously rather than reflecting new knowledge- ex: TBL

3) interval

3) uses continuum of numeric values known as continous data - values have meaning and the intervals are equal - specific meanins are applied to the distances between categories - on interval scales, the 0 point is arbitrary and not absolute- NO ABSOLUTE ZERO! ex) celsius scale, 0 does not mean absence of temperature ex) intelligence measures, personality measures, and manual muscle testing , SAT, ACT, TOEFL

4) ratio

4) data possesses characteristics of interval level of measurement - there IS absolute zero and the comparison can be made in ratio ex) age, height, weight, income

5) instrumentation 6) mortality

5) changes made in the way variables are measured- ex: measuring BP and first using a manual device but sometimes its collected using a digital device 6) the loss of subjects before teh study is completed- r/t a difference in characteristics of the subjects who dropped out compared to those who completed the study

types of classic experimental designs

6 diffferent types

1. Pre-experiments

- a posttest only design that involves manipulation of the IV but lacks control for extraneous variables - O,X,O - give breastfeeding mother education then a posttest, but without giving them a pre-test you can't tell what their prior knowledge was

communnity based participatory action

- based off of the philosophy that when the users of research are involved from the start of the research process, findings that are practical and relevant to the community needs are more likely to result

(probability sampling) systematic random sampling

- every xth element is selected from a numbered list of all elements in the accessible population- the starting point on the list is randomly selected x= size of sampling frame/size of sample

strengths of experimental research limitations of experimental research

- good way to do them if you cant do it via experiment, efficient and effective way to collect data w/o a problem - cant manipulate variables or assing people to control/tx groups

translational research

- research for the purpose of linking research findings to the point of care - determining the best way to get the findings in research to people

2. two-group posttest-only design

- researchers randomly assign subjects to interventino or control group- after intervention is completed the DV is measured at same point in time for both groups ex: looking at post-op pain -randomization, treatment, obervation

3. time series designs

- used when neither randomization nor a comparison group is possible - opt to study one group over a long period of time prior to admin the intervention then make multiple observations - O, O, O, X, O, O, O

2. non-equivalent control group pre-test post-test

-differs from classic experimental design bc researchers are unable to randomly assign subject groups - nonintervention group is COMPARISON GROUP not CONTROL GROUP ex: giving a pre-test to a whole group of students, intervention (teaching in class), posttest (final) next semester they do giving a pre-test to a whole group of students, intervention (teaching online), posttest (final)

types of experimental research designs

-origin: level III research questions: asks "what are the effects of ___ on ____" -basic idea: trying to explore cause and effect properties of a true experiment: manipulation, control, randomization

6 components of the research design....

...

CHAP 13

CHAP 13

match the terms a) nominal b) ordinal c) interval d) ratio

a) types of cancers b) strength of contractions (mild, moderate, strong) c) pH of urine d) intake and output measures

study validity

ability to accept results as logical, reasonable, and justifiable based on the evidence presented 1) internal validity 2) external validity

control

ability to manipulate, regulate, or statistically adjust for the multitude of factors that can influence the DV - necessary to make assertions about cause and effect - how can a nurse researcher achieve control? 1) allow no variation 3) specify the variation to be allowed 3) distributing the variation equally- through randomization

(nonprobability sampling methods) convenience sampling

accidental sampling0 select elements for inclusion in the sample bc they are easy to access

4. solomon four group design

an experimental design with four groups- some receive the intervention, some are measured before and after, other sa re measured only after - 2 groups receive the intervention, 2 groups receive the placebo - superior to the two-group-posttest only desi bc selection bias is minimized -but must be able to get large sample size for this method

B2. comparative designs

based on level II research i. features: a research design that involves comparing and contrasting 2 or more samples on one or more variables, often at a single point in time ii. based on level II research questions: what is the difference... iii. aim iv. issues- sampling, observations and timing model testing design: all paths expressing relationships between concepts and identified, and a conceptual map is developed- analysis determines whether or not the data are consistent with the model

B1. correlational design characteristcis

based on level II research questions, "what is or what are the relationships between/among variables?" i. features --> research design that relates multiple varibales measured at a single point in time in a sample from a designated population ii. aim iii. issues --> sampling observations and timing

non experimental

can be used for the purposes of describing a phenomenon in detail, explaining relationships and differences among varibales, and predicting relationships and differences among varibales

Experimental designs have control groups- quasi experimental designs have which of the following?

comparison groups

5) interaction of tx and history

concerned with how the effects from the intervention might be changed by events occurring in the past or in the future- ex: found that comp based intervention to inc condom used- are those results generalizable to the future if a cure for HIV were to be discovered

univariate analysis

conducted to present organized information about only one variable at a time and includes info regarding frequency, distributions, measures of central tendency, shape of the distribution, and measures of variability

1. two-group pretest posttest design

considered to be the classic experimental design - subjects are randomized to either the experimental group receiving the intervention or the control group- they are measured before and after intervention is given - randomization, observation, and treatment, observation

what is the researcher's ability to manipulate or regulate extraneous variables known as?

control

elements of a good design

controlling the enviro, controlling equivalence of subjects and groups, controlling the treatment, controlling measurement, controlling extraneous variables

B. quantitative designs for exploring association and difference

correlation and comparative designs

descriptive statistics

deal with the collection and presentation of data used to explain characteristics of variables found in a sample - describe, summarize, and synthesize collected data - this information must be accurate

internal validity

degree to which one can conclude that it was the IV NOT EXTRANEOUS VARIABLES that produced the change in the DV - researchers must demonstrate that results obtained were caused by the IV - common treats to internal validity= bias, history, maturation, testing, instrumentation, mortality, and statistical conclusion validity

to describe the frequency of the single variable myocardial infarction in adults ages 30-49, which of the following could be used?

descriptive statistics, univariate analysis

ex of descriptive vs. inferential stats - efficiency of vitamin C in preventing the common cold

descriptive stats --> used to report that 60% of subjects in the experimental groups had fewer colds than did subjects receiving the placebo inferential stats --> determine whether the difference in the number of colds between the 2 groups was statisticallly significant

A1. descriptive designs

designs that provide a picture of a situation as it naturally happening w/o manipulation of variables

quantitative variables can be discrete or continuous

discrete variables- values that are countable, but do not assume numeric value between countable categories continuous variables- variables that have every possible value on a continuum

(nonprobability sampling methods)

do not require random selection of elements and therefore are less likely to be rep of the target population

randomization

effective way to control extraneous variables - can occur either with random sampling of the subjects to be studied or by random assignment of subjects to intervention or control group

which of the following is NOT a purpose of nonexperimental designs

examine causality

3. factorial design

experimental design using 2 or more experimnetal groups with 1 control group

5. crossover designs

experimental designs that use 2 or more treatment, subjects receive treatment in random order

what is the degree to which the results of studies can be generalized to other individuals, settings, or time?

external validity

in factorial design, only 1 group of subjects is required because they act as their own control

false (thats true in a crossover design)

a researcher plans to observe children in a kindergarten class- students have always been told to be on their best beahvior when guests are present in the classroom- what is the greatest threat to external validity

hawthorn effect

during a study examining nurses' job satisfaction, the union decides to hold a strike. this is what type of threat to internal validity

history

(probability sampling) randomization

important because it reduces the treat of selection bias

inferential statistics

involve analysis of data as the basis for predictions related to the phenomenon of interest - used to make inferences or draw conclusions about a population based on a sample - *they are used to develop population parameters from sample statistics*

(probability sampling) simple random sampling

involves ranodmly selecting elements from the accessible population and is considered by some authorities to be the most effective method to obtain a representative sample - time consuming for the researcher, need a larger sample size to ensure that population is adeuately rep

(probability sampling) sampling frame

list of all possible elements in the accessible population

qualitative research

lowest level!!! level I- "what is x" - we do level I research when we dont have a lot of previous research so we have to start very basic - whats the magnitude/level of the problem

quasi experimental designs include which of the following essentail components

manipulation of the IV (NO RANDOMIZATION)

3) setting, the "where"

naturalistic or lab setting- lab is very well controlled, human subjects have equipment/physical errorness

levels of measurement (pg 271)

nominal ordinal interval ratio

representativness

obtaining representative samples so that results of studies can be generalized to target populations - generalizability, aka external validity, is the applicability of study findings to target populations

rank the evidence generated from the following designs from lowest to highest

one group posttest only, nonequivalent groups posttest only, nonequivalent control group pretest-postttest, experimental design (highest)

researchers can use several sampling methods to select subjects- divided into 2 categories

probability nonprobability

A. quantitative designs for describing the incidence or prevalence of a phenomenon --> aka descriptive study designs

quantitative descriptive designs focus on "what is the magnitude of the problem, how common is the phenomenon?"

when a researcher assigns subjects to groups by tossing a coin, the researcher is using which technique?

random assignment (all subjects have an equal chance of being assigned to treatment or control group- ex: randomly assigning first person who comes through the door- one person is tx group and one person is control) random sampling (not the answer to the question- but means that all ppl in the population of interest have the same probability of being selected to be included in the study)

(probability sampling) cluster sampling

random sampling method of selecting elements from larger to smaller subsets of an accessible population- multistaging sampling

manipulation

refers to the ability of researchers to control the IV ex) manipulation of the educational intervention and compare differences

external validity

refers to the degree to which the results of the study can be generalized to other subjects, settings, and times threats to external validity: construct validity, effects of selection, interaction of tx and selection of subjects, interaction of tx and setting, interaction of tx and history

causality

refers to the relationship that exists between a cause and its effect - the cause has the ability or power to produce a specific effect - to determine causality- the varibales of interest are studied starting with 1) casual variable (IV) has to occur prior to the occurrence of the DV - use experimental designs to determine causality! 3 conditions... 1) there must be a strong correlation between proposed cause and effect 2) proposed cause must precede the effect in time 3) the cause has to present when the effect occurs

quasiexperimental designs

research designs involving the manipulation of the IV but lacking either random assignment to groups or control group (3types)

when designing a study, which of the following should the researcher consider?

research question, review of the lit, theoretical framework, study purpose

probability

research related to human health and functioning results in assertions of probability or how likely it is that the change in the DV was caused by the IV - leave the possibility that there were other influencing factors on the change in DV

snowball-

researchers assume that individuals will identify with others who are similar to them

bias

results when extraneous varibles influence and distort the relationship between the IV and the DV so that the findings are not really reflecting the true relationships ex) if the study included students w/o a current sexual partner- it would be highly likely that the intervention would result in no significant difference in condom use- control for bias by carefully selecting participants

order of relationship of sample to population

sample --> accessible pop --> target pop --> population

to use random sampling, a researcher must know the

sampling frame

(probability sampling) probability sampling

sampling method in which elements in the accessible population have an equal chance of being selected for inclusion in the study

(probability sampling) stratified random sampling

selecting elements from an accessible population that has been divided into groups or strata -each element can be only put into one stratum - ex: stratum according to gender- subjects can be cateogried only as boys are girls - reduced sampling error

research design

should flow from the stated purpose and provide a plan that can answer the research questions and test the stated hypothesis - blueprint for conducting the study - map to guide the researcher to reach the goal

researcher desires to study the effect of a memory game on older adults abilities to recall a short term task list- researcher randomly selects 20 elders from a list of residents at a local nursing home- sampling method used?

simple random sampling

within- groups design

situations when researchers are more interested in making comparisons within the same subject- comparisons are made about the same subjects at 2 or more points in time or on 2 or more measures ex) researcher is interested in the effect of music therapy on patient's levels of pain- measure the subjects level of pain before the intervention, conduct the intervention, adn then measure pain levels after the intervention - by comparing subjects pain scores before and after intervention- researcher is able to determine effectiveness of music therapy as pain relief

between groups design

studies developed to compare 2 different groups of subjects ex) researcher who is studying condom use among adolescents may wish to know the practices of high school juniors and seniors as well as colllege freshmen and sophomores- make compairosns among these 4 groups about the frequency of condom use

hawthorn effect

subject's beahviors may be affected by personal values or desires to please the experimenter- reactivity behavior can be affected by their personal values, their desires to please the experimenter or provide the results the experimenter wants and congruence of the study with subject's personal interests and goals

which of the following sampling methods involves randomization?

systematic sampling, stratified random sampling, cluster sampling

7) statistical conclusion validity

the degree that the results of the statistical analysis reflect the true relationship among the IV and DVs

major difference between experimental and non-experimental

the role of the researcher experimental --> researchers actively manipulate the IV to determine the DV- involve randomization and the use of control group whose results can be compared to the group receiving the intervention nonexperimental -->researchers are observers noting the occurrence of the varibales of interest and trying to determine relationships and differences

multivariate analysis

the use of statistics to describe the relationships among 3 or more variables

bivariate analysis

the use of stats to describe the relationship between 2 variables

extraneous varibales

those that confound, confuse, the effect of the IV on the DV - researchers can control for extraneous variables through careful selection of participants, use of consistent data collection procedures, randomization, or use of certain statistical tests ex) age of the teen, religious beliefs, self efficacy for resisting peer pressure, cost of condoms

quantitative designs are used...

to examine relationships among variables and are categorized as experimental or nonexperimental

purpose

to fill a gap in knowledge with the research questions or hypothesis determining what specific new knowledge will be generated

(nonprobability sampling methods) purposive sampling

to select a distinct group of individuals who either have lived the experience or have expertise in the event or experience being studied- sampling method to recruit specific persons who could provide inside information - use snowball sampling or network sampling

patients who are not in the intervention group must receive the usual standard of care

true

the solomon four group design is more effective at controlling for the threat of testing than is the 2 group pretest-posttest

true

experimental designs

used for the purpose of examining causality

specific uses for quantitative designs

used in translational research, community based participatory action research, and health services research

qualitative variables

values that are non-numeric such as systolic blood pressure, measured in categories such as high, normal, low

quantitative variables

values that are numeric such as systolic blood pressure (120mm/Hg)

2) observations of variables, the "what"

whats the phenomenon that you're studying? pain, HR, etc

type II error

when researchers inaccurately conclude that there is no relationship among the IV and DV when an actual relationship does exist- researcher accepts null hypothesis but they should have accepted it - often times due to a small sample size

(nonprobability sampling methods) quota sampling

when researchers sample from strata w/o randomly selecting elements- using quota sampling - but difference= the use of random selection, in quota data is sampled convinently not randomly

sampling bias

when the sample includes elements that over-or underrepresent characteristics when compared to elements in the target population - this is a threat to the external validity of the study

1) description of the participants involved, the "who"

who are the units of analysis, could be groups of people, communities, individuals, some with certain types of problems, groups whoa re different generations

1) researchers should design studies that are easy for subjects to participate in regardless of how much control over extreneous variables is achieved? 2) nursing body of knowledge is dependent on the quality of research findings that are disseminated

1) F 2) T

1) in retrospective designs, aka ex post facto designs, the researcher manipulates the IV 2) cohort comparison studied can save time because more than one group of subjects is studied 3) the treat of mortality is greater in cross sectional designs than in longitudinal designs 4) any study that involves collecting data at multiple points in time is a longitudinal study

1) F 2) T 3) F 4) F

1) construct validity 2) effects of selection

1) a threat to validity when the instrument does not accurately measure the theoretical concepts- 2) threats to external validity when the sample does not represent the population - ex: interviewing mothers during the day- but some mothers may be doing childcare- etc

purpose of non-experimental designs

1) describe phenomena 2) explore associations and differences among variables 3) predict relationships

when choosing research design think about

1) purpose of research design 2) level of research question 3) 6 components of the research design

essential components of experimental designs

1) randomization 2) control 3) manipulation

4) measures of time, the "when"

1) retrospective designs: research designs when researchers look back in time to determine possible causative factors- aka ex post facto- ex: start with the DV and look back to see what caused it, the IV cannot be manipulated since it is back in time 1a) case control- type of retrospective study in which researchers begin with a group of ppl who already had the disease- studies that compare 2 groups: those who have a specific condition and those who dont have that condition 2) cross-sectional designs: nonexperimental designs that researchers use to gather data from a group of subjects at only one point in time- snapshot by collecting IV and DV at the same time- ex: study design to measure exposure and disease as each exists in a population or rep sample at one specific point in time 2b) cohort comparison- nonexperimental cross-sectional design in which more than one group is studied at the same time so that conclusions about a variable over time can be drawn without spending as much time, ex: gathering data from different subjects at each grade saves time rather than waiting 4 years to collect the data 3) longitudinal designs- used to gather data about subjects at more than one point in time- look at present and end in the future 3a) panel designs: the same subjects are used to provide data at multiple points in time 3b) crossover designs: subjects recieve more than one experiemntal treatment and are then followed over time- subejcts act as their own control group- manipulate the IV by randomizing the order in which the tx are provided

1) nominal

1) the lowest level of measurement whereby data are categorized simply into groups- categorical data- WEAKEST LEVEL ex) responses of "yes" and "no" on a survey are assigned numbers 1 and 2- the value of tehse numbers have no meaning bc one cannot say tha ta "no" response if higher than a "yes" response - COMMONLY FOUND IN QUESTIONNAIRES- closed ended questions about gender, race, and diagnosis

developing the study design involves 6 steps

1) what is the purpose of the study? 2) does the study focus on determining the effects of an intervention (treatment)? 3) what is the number of groups of subjects in the study? 4) what is the number of times that the DV is measured and the time of these measures 5) what types of comparisons of measure of the dependent variable will be made? 6) what strategies for controlling extraneous variables in the research setting and among the study's subjects will be used?

1) selection bias 2) history

1) when the change in the DV is a result of differences in the characteristics of subjects before the entered a study rather than a result of the IV- MIN by use of random assignment 2) when the DV may have been influenced by some event other than the IV that occurred during the course of teh study- cant attribute inc condom use to your edu if at the same time MTV was promoting condom use - MIN by including a control group