Research EXAM #3 Flashcards

(91 cards)

1
Q

Type 1 vs. Type 2 error

A

Type I — false (+) if investigator rejects null hypothesis that is actually true in the population
Type II — false (-) if investigator fails to reject a null hypothesis that is actually false in the population

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Statistical tests to improve validity

A

Determine probability of errors:​

Type I​

Type II​

Calculate and report Effect Size​

Ensure data meets the basic assumptions of the statistical tests​

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Empirical literature

A

this type of literature is based on experiences or observations and displays
how theories apply to individual behavior or observation. E.g. if a medication is going to reduce
BP, then you hypothesize that because the medication reduces BP in will lower the risks for
heart attacks. This notion is based on observation rather than theory

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Seminal literature

A

foundational and classic literature (think o fit as original, first creation of
the concept)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Construct variable

A

in simple terms it refers to an explanatory variable that cannot be directly
observed. We may be able to observe its effect but not a directly observe it. A good example is
gravity, we know a lot about what gravity does and its effect but we cannot directly see or
observe gravity. So gravity maybe considered a construct variable. Anxiety is another construct
variable. We can observe the outcomes of anxiety such as tension, stress, anger etc.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Correlational (relationship) design

A

demonstrates a relationship between the variables it may
be a positive relationship or negative relationship

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Common Types of Statistical Tests (Inferential tests)

A
  1. Independent t-tests: tests for differences in the means between 2 groups
  2. Paired t-tests: tests for differences between paired measurements e.g. pretest before
    the intervention and post-test after the intervention
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Analysis of variance (ANOVA)

A

F-test - is a statistical test used in experimental or quasi-
experimental research (quantitative). Test is applicable when measuring the means between 2
or more study groups. Only measures one dependent variable. Example in a diabetic study one
group receives oral medication, one group is diet controlled, one group is placebo.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Multivariate analysis of variance (MANOVA)

A

a statistical test used when measuring 2 or
more dependent variables (outcomes) for the research
– e.g does music therapy and relaxation techniques affect anxiety and panic attacks?​

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

ANOVA vs. MANOVA are

A

statistical tests

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What is the purpose of a chi-square test?

A

helps detect the relationship between 2 variables but does not demonstrate
the depth or direction of the relationship between the variables

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Correlational coefficient (r)

A

measures correlation (relationship) between 2 variables.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

External validity

A

Generalize the findings from a research study to other populations, places, situations
— It is SO good that nothing needs to change —> generalization = QUANTitative

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Why is validity important in search?

A

It helps to measure what it is supposed to measure

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Internal validity

A

Ensures the intervention worked and the outcome was not based on other causes
—The confidence experimental treatment/condition has made a difference and rival explanations were systemically ruled out through study design and control
— IV caused the outcome (DV)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Generalization is synoymous with ____.
Transferability is synonymous with ____.

A

QUANTitative = generalization
QUALitative = transferability

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

Instrumentation

A

The instrument or data collection process has changed
— e.g. operative error: post-partum tool and how explained vs. someone who is not well-informed, may not use the tool/instrument to the best of its abilities

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

Describe some biases that can be interjected into a research study

A

— Personal
— Selection
— Subject

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

Factors that can interfere with internal validity

A

– Historical threats
– Maturation
– Testing
– Instrumentation
– Consent effect
– Treatment effect
– Hawthorne effect
– Multiple-treatment effect
– Subject selection
– Attrition

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

Describe a historical threat

A

subjects behave in a certain manner because of their exposure to events outside the experiment
– e.g. the occurrence of an actual earthquake during a field study of the effects of training in earthquake preparedness

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

Bias in research

A

Sampling error:​

A number that demonstrates the difference in the results between the sample and the population it was drawn from​

Treatment effects:​

A threat to internal validity because the subjects may perform differently​

Measurement error:​

Difference between actual attribute (true score) and the amount of attribute that was represented (observed score)​

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

Maturation related to internal validity

A

Too much time has passed with long study
Changes in the research subjects not due to the intervention, because time has passed

— e.g. deterioration of physical characteristics: vision, hearing, taste, memory
— e.g. pain relief w/ cancer patient disease is maturing (stage I —> stage III)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

Testing related to internal validty

A

Familiarity of the research subject w/ testing, especially retesting occurs
— e.g. pre-test then take post-test; knowledge became enhanced

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

Consent effect

A

Threat to internal validity occurs. Because the subject who consents to student may differ from those who do not in way that affects outcome of the study
— e.g. change your mind to study after finding out you’re going to be recorded therefore change their mindset/thoughts

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Treatment effect
Subject may perform differently b/c they know they are being in a study therefore act the way that looks good — e.g. masks their behavior b/c if in an anger study, you will hide that anger
26
How to control extraneous variables
-- Eliminate the threat -- Control the threat -- Account for the threat or write-up
27
Hawthorne effect
Seen as person of power and subject takes on subordinate role; research subjects change their behavior in a study because they are aware of being under observation — e.g. You are researching the smoking rates among bank employees as part of a smoking cessation program. You collect your data by watching the employees during their work breaks. If employees are aware that you are observing them, this can affect your study's results
28
Multiple-treatment effect
An inability to isolate the effects of treatments because multiple treatments are being used at the same timeA
29
Subject selection/Subject effects (bias)
A threat to internal validity due to introduction of bias through selection or composition of comparing groups assignment of research subjects to groups in a bias manner, not random - e.g. Health studies that recruit participants directly from clinics miss all the cases who don’t attend those clinics or seek care during the study.
30
Attrition
Loss of research subjects during the study (possibly due to how long the study is); drop out of study — e.g. children from troubled families increase the likelihood to drop out/move away in the middle of school year — e.g. yoga education vs. paper education over 8 weeks —> only 1 showed up/completed the paper education study vs. everyone that completed the yoga education
31
What factors can affect *external* validity?
-- Population -- Ecological -- Time & historical effects -- Novelty -- Experimenter effect
32
Threats to external validity
— Selection effects — Time — History — Novelty effects — Experimenter effects
33
Threats to trustworthiness in QUALitative research
— Hawthorne effect — Selection effects — Historical effects — Researcher bias
34
Researcher bias
The primary investigator (PI) poses their thoughts into the study
35
Population validity
Can the findings be generalized from the sample to a larger group? The capacity to confidently generalize the results of a study from one group of subjects to another population group — e.g. can the findings of patients from one medical center be generalized to patients in an entire health care system?
36
Ecological validity
*external validity* Findings can be generalized and applied to other settings Can findings be generalized from one set of environmental conditions to another? — e.g. Can the findings be generalized from a medical-surgical unit to a long-term unit.
37
Time & historical effects r/t external validity
Threats to external validity
38
Novelty effect
*threat to external validity* results based on the newness of the treatment rather than actual procedure - hard to decipher what caused the outcome Occurs when subject reacts to something b/c its novel/new, rather than actual treatment or intervention itself — e.g.
39
Experimenter effect
*threat to external validity* Results may reflect personal biases of the researcher Interaction w/ researcher conducting the3 study or applying intervention — e.g.
40
These improve validity of qualitative research
-- Verbatim accounts -- Triangulation -- Bracketing -- Audit trials -- Member checking
41
Verbatim accounts
“Word-for-word” statement of a response *NOTE: Improves validity of QUALitative research*
42
Triangulation
A means of enhancing credibility by cross-checking information and conclusions using multiple data sources using multiple methods or researching to ensure credibility — The study, phenomenon using multiple theories/perspectives to help with data *NOTE: improve validity of QUALitative research*
43
Bracketing
A researcher separates own experiences from what is being observed in the study, to reduce bias A method of limiting the effects of researcher bias and setting them aside by demonstrating awareness of potential suppression of research *NOTE: improve validity of QUALitative research*
44
Audit trials
establish findings are based on participant’s responses not researcher
45
Member checking
data is sent back to participants to check for accuracy Having participants review and comment on accuracy of transcripts, interpretation or conclusions — e.g.
46
What is the purpose of descriptive research?
Actively describes population, situation, or phenomenon — the what, when, where, why, how can be an answer
47
Name some descriptive type of research studies
— Case reports — Case — Cross-section — Sectional — Ecological study
48
Longitudinal study
Studies conducted by following subjects over a period of time w/ data collection occurring at prescribed intervals
49
Cross-sectional study
Study conducted by looking at single phenomenon across multiple populations at a single point in time w/ no follow-up design -- e.g. a medical study examining the prevalence of cancer amongst a defined population. The researcher can evaluate people of different ages, ethnicities, geographical locations, and social backgrounds.
50
Case study
Exploration of single unit of study such as person/family group or community
51
Single-subject
Uses a single case/subject in which baseline data are collected, an intervention is applied, and responses are tracked over time
52
Epidemiology
Branch of medicine dealing w/ incidence, distribution, possible control of diseases + other factors r/t health
53
Pilot study
“Feasibility” = small-scale preliminary study — e.g. “live & learn”/assess planned study’s cost, time, efficiency, accuracy on smaller scale; see if there are any problems — make sure instrument is gathering data that you need to
54
Name the different types of methodology used in research
1) Simple 2) Systematic 3) Stratified 4) Cluster
55
Difference b/w experimental design and quasi-experimental design
56
Experimentation
Use scientific methods to establish the **cause and effect** relationships among a group of variables that make up a study — e.g. clinical trial where participants placed in control and treatment groups —> determine the degree which interventions in the treatment group is effective or not
57
Quasi-experimental
Attempts to establish cause-effect relationships among groups of variables that make up a study — QUANTitative research method — # data collection an statistical analysis e.g. You hypothesize a new after-school program will lead to higher grades — choose 2 similar groups of children who attend different schools —> implements new program —> not implementation
58
Level of measurements - what are they and examples, descriptive designs the different types and what they mean, central tendency what is it examples.
59
internal and external validity, how do you maintain this in research, what does it mean?
60
longitudinal, predictive, single-study, correlation, survey -- what are they?
-- Predictive: -- Single-study: -- Correlation: -- Survey:
61
ways qualitative researchers maintain accuracy in in their research?
62
generalizability, transferability, experimental, quasi-experimental, gold standard.
63
Examples of descriptive research question and designs
-- Descriptive: What is nurses' knowledge about best practices r/t to oral care? -- Survey: How are nurses involved in decision making about patient care, the work environment, and organizational practices? -- Cross-sectional: What are the differences in job satisfaction among nurses at different stages of their careers? -- Longitudinal: What is the effect of urinary incontinence on the quality of life of long-term-care residents over time? -- Case study: What are the appropriate assessments and interventions for a patient experiencing paraplegia after heart surgery? -- Single-subject study: What were the responses of an individual w/ type 2 DM to culturally appropriate counseling from a nurse? -- Correlation: What is the relationship b/w patient satisfaction and the timeliness and effectiveness of pain relief in a fast-track emergency unit? -- Predictive: Can feeding performance in neonates be predicted by indicators of feeding readiness?
64
Central tendencies
CT - single number used to summarize the data​ Mean (average)​ Mode (number occurring the most in the data set)​ Median (middle number in the data set)​ ​
65
Derived variable is not the primary variable
Not IV or DV that you study — e.g. can derive BMI from height and weight
66
Value of descriptive data
Descriptive data are summarized and analyzed Provide useful information about the research participants Offers a quick glance Central tendency - measures how variables are alike Variability - measures how variables are different Helps the reader understand the information — It summarizes and analyzes
67
1st step of analyzing data in QUANTitative research, how is the information gathered
From surverys
68
Examples of descriptive data
Derived from the data set: Age of research participants (data set of34 males, 35 females) Years of work experience Scores on job satisfaction
69
Nominal data
*also known as category* — e.g. colors, male or female, black or white, BSN students vs. PA students
70
Ordinal data
Categorical data that are ranked: Pain scale has numbers from 1 to 10. On the scale 5 > 3, 10 > 8. We can’t tell that the difference between the two sets even though they are on the same scale
71
Categorical Data examples vs. Continuous data
_Categorical:_ — Nominal = BSN vs. PA students — Ordinal = pain scale _Continuous:_ — Interval = temperature (NO TRUE ZERO) — Ratio = height, length, age (USES TRUE ZERO AND GOES UP)
72
4 levels of measurement
Ratio Interval Ordinal Nominal
73
Levels of measurement: categorical vs. continuous
Nominal Categorical data: leaders/nonleaders, males/females, ADN students/BSN students Ordinal Categorical data that are ranked: Pain scale has numbers from 1 to 10. On the scale 5 > 3, 10 > 8. We can’t tell that the difference between the two sets even though they are on the same scale Interval Data ranked with equal intervals e.g. temperature (don’t have a true zero, you can measure temperature below zero) Ratio Interval data that have a true zero e.g. height, length, age
74
EXAM IMPORTANT INFORMATION
— Scatter plot — Measurement of height is considered = ratio — What is the 1st step in analyzing quantitative research data? Summarize descriptive data — Standard normal distribution: mean = 0 and std. deviation = 1 — Measures of central tendencies = mean, median, mode — A measure that is consistent, biased, and not accurate = systematic — Median in their data — Know mean, median, mode, range — Suppressive variable/confounded/extraneous — Ratio, ordinal, nominal
75
Scatterplot’s purpose
Indicates the nature of the relationship between 2 variables - strength and direction
76
Purpose of box and whiskers plot
Shows distribution over time Box contains 25th to 75th percentiles of the distribution
77
Difference b/w positive z-score vs negative z-score
– **z-score** depicts the number of standard deviations above or below the mean an observation falls –Positive z-score: Above the mean –Negative z-score: Below the mean
78
Standard scores
Describe the relative position of an observation within a distribution
79
A priori
The researcher has to lay out EVERYTHING in front before conducting the study (everything they’re doing, methodology, design, sample sizes, location of sample sizes, laid out) —> IRB and come back —> CANNOT change b/c jeopardize subject (even if they thought of something better that can impact the study)
80
Lack of disclosure
They don’t tell because it’s going to change the behavior; allowed b/c the details they leave out will not be details that cause harm/death to participants — e.g. observe if you’re friendly with each other (in a psych facility and outdoors
81
Common statistical errors
*Using inappropriate statistic for level of measurement *Incorrectly entered data *Data presented without context *Lack of disclosure *Overinterpretation of results *Inconsistent or misleading presentation — must have enough details to judge the findings
82
Evaluating descriptive data
*Enough details are provided to judge *Presentation is understandable and clear *Complete statistics are provided for each variable *Graphical representations are consistent *Statistics are not misleading, misinterpreted, or confusing
83
How do we use descriptive data?
*Check trends in patient status *Review quality performance *Analyze distribution of disease in populations *Develop evidence about clinical outcomes *Use as basis for designing an intervention study *Judge patient conditions and make improvements
84
Descriptive data is important when related to
Age and gender
85
Prevalence vs. incidence
Prevalence: # of existing cases — e.g. # of diseases in a population Incidence: # of NEW cases
86
Retrospective studies
A study that compares two groups of people: those with the disease or condition under study (cases) and a very similar group of people who do not have the disease or condition (controls). -- e.g. a group of 100 people with AIDS might be asked about their lifestyle choices and medical history in order to study the origins of the disease.
87
Prospective studies
88
Methodology refers to which method(s) in a study?
— Quantitative method — Qualitative — Mixed method
89
Does not use randomized groups
90
Range
The highest and lowest values in the data set
91
Researcher attempts to show the spread of the data from the mean. Which of the following refers to the spread?
Variance