EIP final review
docx
keyboard_arrow_up
School
Texas A&M University *
*We aren’t endorsed by this school
Course
1807
Subject
Health Science
Date
Jun 24, 2024
Type
docx
Pages
7
Uploaded by MegaStarLemur35
Final exam review
UNIT 9
Qualitative research
To identify, describe, and ultimately understand human/ organizational behaviors, attitudes, and experiences and how they influence health
Uses the subjects’ own words and narrative summaries of observable behavior to express data NOT NUMBERS
Relies on words and images
PROS
: Smaller sample size, open-ended process, targeted, specific, more affordable
CONS
: Time consuming, subjective data, limited generalizability due to small sample size, data can be missed if the wrong questions are asked
No numbers
Descriptive
Rich detail Thematic analysis
Qualitative research five main design
Case study
Followed 1 person over time- ex: patient w a rare diagnosis and they followed them
Interviews, Documents, Reports, Observations Narrative
Time-consuming, translated videos into a narrative
Phenomenology
Interest that involves more than 1 person, interview and sit down to get information
Grounded theory
The researcher doesn't have a theory in mind, not well studied or evidence to support a theory
Ethnography
Cultures, context, how do people differ, or share routine/rituals Interest in the culture within the person
Qualitative sampling methods
Purposive sampling
Subjects aren’t selected randomly but rather a specific purpose by the researcher to best meet the needs of the study “judgment sampling’
Convenience sampling
Obtained by ‘convenience’ from a group of people that are easy to contact
Voluntary response
Easy to contact/ participants volunteer to be in
Snowball
When it’s difficult to recruit, researchers will use participants to recruit others, as they get in contact with more people
How do you collect qualitative data (qualitative data collection methods)?
Nonprobability sampling methods where subjects/individuals are selected using non-random criteria, not every person has a chance of being included
UNIT 10
Mixed Methods studies
Using both quantitative and qualitative approaches within 1 study strengthens the validity of the findings
Concluding through different routines qualitative and quantitative
Convergent
Exploratory sequential
Explanatory sequential
Literature review as a process vs a product
Process identifies topic, searches literature, selects/organizes articles, develops thesis/purpose
Product summarizes articles on a topic
Differentiate between the different types
Narrative, systematic review, scoping review
UNIT 11
Reliability (repeatability)
Can you reply on the results?
Formula= true score/ observed score
Measures of reliability
What is Internal consistency
? Be prepared to judge whether a scale has excellent, great, good, or poor internal consistency
What is inter-rater reliability
and how is it measured? What % agreement between the two raters is considered good?
The same phenomenon is measured by DIFFERENT observers 10-20% of your data and have the 2
nd
person independently code it
What is intra-rater reliability
and how is it measured? What % agreement between the two raters
is considered good?
The same person measures the same phenomenon REPEATEDLY Variability (measures what it's suppose to measure)
Content Do two measures agree on the same result?
The results are supported by another scale
Criterion (concurrent & predictive) Does one outcome predict the result of another?
Construct Does the tool measure a theoretical construct accurately?
Does the results from our muscle strength scale correlate with scores from a ROM scale?
QUALITATIVE (CATEGORICAL VARIABLES)
Nominal (by name) gender, race, eye color
Ordinal (in order) letter grade, agree/disagree
QUANTITATIVE (NUMERICAL VARIABLES)
Interval (distance is meaningful) temperature, 1-10 rating scales
Ratio (absolute zero) time, weight, height
Pg. 266
Range:
difference between highest and lowest values in the distribution
Variance:
considers the dispersion of individual values around the mean
Standard deviation:
derived from variance and allows variance to be expressed in the same unit of measurement
Z-SCORE
Null hypothesis: NO SIGNIFICANT DIFFERENCE between specific populations, any observed difference being due to sampling/ experimental error
Cohen's d value- effect size
Small effect size d= .2 or less
Medium effect size= .5
Large effect size= .8 or more
Alpha 95% or .05
Type 1 error: difference in statistical conclusion
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
- Access to all documents
- Unlimited textbook solutions
- 24/7 expert homework help
no difference in groups
Type 2 error: no difference in statistical conclusion
difference in groups
How do you choose a statistical test? Based on level of measurement for DV & experimental design (nominal, ordinal, ratio)
Degrees of freedom represent: # of items that are free to fluctuate n-1 Test
Study Design
Example
Level of
measurement
Test
Family
Independent t-test
Differences between two groups Measuring the increase in weight following a therapeutic diet between Group A and Group B
Reported as:
t
(degrees of freedom) = the t
statistic, p
= p value.
Group A gained more weight (M = 83, SD = 5) than Group B (M = 30, SD = 5), t(48) = 2.3, p = .026.
Numerical Para-
metric
Paired samples t-
test
Within group differences (e.g., pre/post test, compare treatment A & treatment B in the same group)
Measuring the increase in weight in Group A before and after the diet
Reported as:
t
(degrees of freedom) = the t
statistic, p
= p value.
There was a significant increase in weight after the treatment Numerical
(M = 83, SD = 5) compared to before the treatment (M = 30, SD = 5), t(28) = 2.3, p = .026.
One-way ANOVA
Differences between more than 2 groups
Measuring the increase in weight following a therapeutic diet between Group A, Group B, & Group C
Reported as:
F
(df1, df2) = the F
statistic, p
= p value.
An analysis of variance showed that the effect of treatment was significant, F(2,49) = 5.94, p = .007. Post
hoc analyses indicated that the average weight gain was
significantly higher in Group A compred to the other two groups. Numerical
Two-way ANOVA
Differences between groups based on 2 IVs
Measuring the increase in weight following a therapeutic diet between Group A, Group B, & Group C (IV #1) based on the number of meals/day (IV #2)
Numerical
Repeated Measures ANOVA
Within group differences (e.g., pre/post, compare treatment A & treatment B in the same group) for more than 1 group
Measuring the increase in weight in Group A & Group B before and after the diet
Numerical
Pearson R
Correlation between IV and DV
Measuring the correlation between weight gain and improved quality of life, as measured by as QoL scale. Reported as:
r
(degrees of freedom) = the r
coefficient, p
= p value.
There was a largesignificant negativecorrelation between weight gain and the QoL scores, r(48) = - .7, p = .026.
Numerical
Mann Whitney Differences between two groups
Measuring the increase in weight following a therapeutic diet between Group A and Group B;
however, the groups are not equal in size, and Group A is not normally distributed.
Categorical *Or Numerical but the assumptions were not met for
a t-test
Non-
para-
metric
Wilcoxon Rank
Within group differences (pre/post
test differences)
Measuring the increase in weight in Group A before and after the diet; however, the group is not normally distributed
Categorical *Or Numerical but the assumptions were not met for
ANOVA
Chi Square
Differences between two groups
How many people preferred the new diet in Group A vs. Group B? Reported as:
X
2
(degrees of freedom, N) = the X
2
value, p
= p value.
The number of participants who
liked the new diet in Group A was significantly higher than the
number in Group B, X
2
(1, 50) = 10.1, p
= .017.
Categorical (MDC) is defined as the minimal detectable change
that falls outside the measurement error in the score of an instrument used to measure a symptom
.
A minimal clinically important difference
(MCID) is an important concept used to determine whether a medical intervention improves perceived outcomes in patients
. Coding- labels themes that emerge as the data are amassed
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
- Access to all documents
- Unlimited textbook solutions
- 24/7 expert homework help
steps to evaluating research 1: classify the research & variables
2: compare purposes & conclusions
3:Describe design & control elements
4: Identify threats to research
5: place the study in the context of other research
6: evaluate the personal utility of the study