Research Final Exam

Measurement involves . . . .
Assigning numbers to objects to represent the amount of an attribute OR Process of assigning numbers to variables, using a specified SET OF RULES
Rules for measuring variables for Research studies must be _____________.
INVENTED. A process must be used to assign numbers
Under what conditions What method (survey, observation) *Numeric values to be used
What are the Advantages of Measurement?
1. Removes the guess work from gathering information.
2. Objectivity: Can be independently verified by others.
3. Produces Precise information.
4. Language of communication.
Define "Errors of Measurement"
If instrument is not accurate then the measures it produces contains a certain degree of error.
Margin of Error (may not be an acceptable range)
They represent an UNKNOWN QUANTITY & they are VARIABLE.

Obtained score = true score + or - error
Factors contributing to Errors of Measurement:
1. Situational contaminants - scores affected by the conditions under which they are produced
2. Transitory personal factors - scores of an individual may be influenced by temporary personal states
3. Response-set biases - number of relatively enduring characteristics of respondents that can interfere with accurate measures of the target attribute.
4. Administration variations - alterations in the methods of collecting data from one subject to the next. FOLLOW PROTOCOL
5. Instrument clarity - if directions for obtaining measures are vague or poorly understood, then scores may be affected. KNOW EQUIPMENT
6. Item sampling - errors are sometimes introduced as a result of the sampling of items to measure an attribute.
7.Instrument format - technical characteristics of an instrument can influence the obtained measurement.
(Open-ended questions may produce different information than Closed-ended questions).
Reliability is the . . . . .
degree of consistency with which an instrument measures the attribute it is supposed to be measuring.
Reliability = stability, consistency, dependability of a measuring tool, can reproduce consistency.

The LESS VARIATION an instrument produces the HIGHER ITS RELIABILITY
The 3 key aspects of Reliability are . . .
1. Stability (Test Re-Test procedures,reliability coefficient).Concern that instrument yields the same result.
2. Internal Consistency - 2 or more trained observers, watching same event simultaneously, independently recording variables in a coding system (Cronbach's Alpha or Kuder-Richardson 20)
3. Equivalence (comparing 2 versions of same equipment or 2 observers - measuring same event) When a consensus measure capturing interrater agreement within a small number of categories is desired . . . .The KAPPA statistic is often used.
Cronbach's Alpha is gold standard for _________. Kuder-Richardson 20 is used for _________.
Cronbach's Alpha = for determining reliability
Kuder-Richardson = for 2 choices (T/F or yes/no)

Better than SPLIT HALF TECHNIQUE because it computes all possible splits to estimate homogeneity
A Reliability Coefficient is based on the computation of a . . . .
Correlation Coefficient, that indicates the MAGNITUDE & DIRECTION of a relationship between 2 variables.
Positive relationship zero to +1.00
Negative relationship -1.00 to zero

Looks for relationship between 2 phenomena
List Advantages/disadvantages to Split-half technique
Adv - Easy to use
Uses just one test - eliminates most of the disadv of test-retest
Disadv - Reliability estimates can be obtained by using different splits (odd/even, first/second half)
Validity refers to the degree to which an . . .
instrument measures what it is supposed to be measuring.
Accuracy of the measure
Face Validity refers to whether the instrument . . .
looks like it is measuring the target construct.
Not strong evidence of validity, Not best way
Content Validity concerns the degree to which an instrument . . .
has an appropriate sample of items for the construct being measured and adequately covers the construct domain. (more common)
Relevant for both affective measures (Psych) & cognitive measures (knowledge).
Based on Judgment of experts
Criterion-Related Validity focuses on the correlation between . . . .
the instrument and an outside criterion.
Criterion must be clear cut & objective.
Predictive Validity is the degree which an instrument can . . . .
accurately forecast the future.
Concurrent Validity is a judgment to the degree to which an instrument can . . .
accurately identify a difference in the present.

Use 2 instruments to measure the SAME concept.
Construct Validity is an instrument's adequacy in measuring . . . .
focal construct, is a hypothesis-testing endeavor.
METHODS: Known groups technique & Factor analysis.
Under Construct Validity, using the Known-groups technique is to know that groups are expected to . . .
. . . differ.
(One group with emphasema and one group that is not.)
Used to CONTRAST scores of groups hypothesized to differ on the attribute.
Under the Construct Validity, using Factor Analysis is a . . .
statistical procedure (mathematical equation) for identifying unitary clusters of items or measures.
Validity is never _______but_________.
Validity is never PROVEN but VERIFIED.
Has to be retested for VALIDITY.
A measuring device that is valid MUST be ______.
An instrument can be reliable without being _____.
Define Assumptions
Beliefs that are held to be true but have not been necessarily been proven OR
A principle that is accepted as being true based on logic or reason, without proof
List 3 types of Assumptions
1. Universal: Assumed to be true by large % of population
2. Derived from a theory of previous research
(stress causes disease)
3. specific to a certain research study: assumptions that prediction is possible

Define Limitations
Uncontrolled variables that may affect study results & limit the generalizability of the findings
Examples: Sample deficiencies, design flaws, weaknesses in data collection
List types of variables
Discrete - has a finite number of values between any 2 points (number of children)

Discrete - Categorical variable - variable with discrete values (gender) rather than values on a continuum (weight)

Discrete - Dichotomous - variable having only two values or categories (gender - male/female)

Continuous - values can be represented on a continuum, Infinite number of values between two points (weight, temp, height)
List the 4 levels of Measurement
1. Nominal - most basic level, variables that are discrete & noncontinuous, Demographic data
2. Ordinal - 2nd Level, variables are assessed incrementally (pain scale, ability to perform ADL's,
worst to best, youngest to oldest)
3. Interval - 3rd level - ranking of order, distance between numeric values, can be measured on a scale (continuous variables)
4. Ratio - 4th & highest level - a scale with an absolute zero in the scale.
Data can always be converted to a _______level but not ______.
LOWER level but not HIGHER
(loss of accuracy & information)
DESCRIPTIVE STATISTICS: Define Frequency Distribution
A systemic arrangement of numeric values from lowest to highest, with a count of the number of times each value was obtained
Numbers can produce a HISTOGRAM
Asymmetrical distribution usually is described as __________.
One tail longer than the other
Longer tail point to RIGHT - Positive skew
Longer tail point to LEFT - Negative skew
A normal distribution or the "bell-shaped curve" is______.
Symmetrical, unimodal
Central Tendency refers to the _____.
Refers to what is in the middle
More representative if values come from center
Define Mode
Number that occurs most frequently
It is quick & easy
Unstable - tends to fluctuate widely within a population
Define Median
Number in the middle, if 2 numbers are in the middle, take the average of the 2 numbers.

Point at which & below which 50% of cases fall.
Referred to as the 50th percentile

Removes extremes & skewed numbers
Define Mean
Most used
The average
The mean is affected by the value of every score
Define Range
Highest score minus the lowest score in a distribution
Easy to compute
Highly unstable, based on only 2 scores
Ignores the variations in scores between the 2 extremes.
Define Standard Deviation
Summarizes the average amount of deviation of values from the mean
Bigger the SD, the more dispersed the scores.
Flatter the Bell curve
Index of the variability of scores in a data set
Define Correlation
Most common method of describing a relationship between two measures
Variables are INTERVAL & RATIO
Pearson's r (correlation coefficient) -1 to +1
Inferential Statistics allow researchers to make inferences about a population based on data from a ________.
They offer a framework for deciding whether the sampling error that results from sampling fluctuations is too high to provide reliable population estimates.
To test the HYPOTHESES about a population
Define Probability
likelihood that something will occur
Always a sampling error
Researcher must decide if sample being used is a good estimate of population parameters
Majority of studies are set @ p<0.5. What does this mean?
that there is less than a 5% chance that results are due to chance. This is the standard
Usually set no higher than 0.5
Can set at <0.1 if study requires
What is Hypothesis Testing?
Through statistical procedures enables researchers to make objective decisions about the validity of their hypothesis.

A process of decision making
What is Null Hypothesis?
States that there is no relationship between research variables, and that any observed relationship is due to chance. Rejection of the null hypothesis lends support to the research hypothesis.
What is a Type I error?
Rejecting the Null when it is true
The lower the p level, the less likely for a Type I error
What is a Type II error?
Not rejecting the Null when there was a significant difference between groups
List ways to decrease the chance of Type II errors
Larger sample size (also for Type I error)
Decrease sources of wide variation (control)
Increase level of Significance
Explain Two Tailed tests
Most often used
Both ends (or tails) of sampling distribution are used to determine the range of improbable values
Define Parametric Tests
involve the estimation of at least one parameter, the use of interval or ratio level data, and assumptions of normally distributed variables

Flexible, Powerful

Preferred for data analysis
Define Non Parametric Tests
are used when the data are nominal or ordinal or when a normal distribution cannot be assumed - especially when samples are small.

No Bell shaped curve

Not as powerful but lower level data used

Sample size can be small
Define t-Test or Student's T
A parametric statistical test for analyzing the difference between two means
Define Analysis of Variance (ANOVA)
A statistical procedure for testing mean differences among three or more groups by comparing variability between groups to variability within groups, yielding an F-ratio statistic (one intervention)
What is Post Hoc analysis
to tell which group had the significance, which group responded to the intervention
What is Multifactor ANOVA (MANOVA)
Two or more independent variables on a dependent variable
What is the Chi square test
A statistical test used in various contexts, often to assess differences in proportions; symbolized as x2

Must have 5 in each group

Used with categories of data

Computed from contingency tables

Differences are calculated based on what occurs from a study compared to what the expected frequencies are
What is Spearman's Rho
A correlation coefficient indicating the magnitude of a relationship between variables measured on the ordinal scale. A NONPARAMETRIC TEST
What does Power mean when referred to research?
Ability of a study to identify relationship or detect real differences among variables.
Why use Power Analysis
It is a procedure used to estimate:
1. Sample size requirements prior to undertaking a study
2. the likelihood of committing a Type II Error
List the 4 Elements of Power Analysis
1. Alpha: probability of making Type I Error, set at 0.5

2.Beta: Probability of making Type II error, should be no more than 4 times the alpha

3. Power is 1 - beta
conventional standard accepted for power is 0.8

4. Effect size - strength of the relationship among study variables (determined by researcher)
Determined from: ROL, Researchers own pilot data, estimates based on clinical experience

With all 4, a sample size can be determined using the Power Analysis formula
As a researcher, you collect data on the marital status of your subjects. The appropriate measure of central tendency to use for this variable is:
A. mean
B. median
C. mode
D. standard deviation
C. mode
The appropriate measure of dispersion for the last question is:
A. Range
B. Variance
C. Standard deviation
D. None of the above
D. None of the above
On what basis does the t-test compare groups?
A. Frequency distributions
B. Mean score
C. Alpha level
D. Confidence Intervals
B. Mean score
Which of the following correlation (Pearson r) coefficients would give you the most precise prediction?
A. r= 0.80
B. r= -0.60
C. r= 0.45
D. r= -0.85
D. r= -0.85 the strongest
The use of inferential statistics permits the research to:
A. Generalize to a population based on information gathered from a sample
B. Describe information obtained from empirical observation
C. Interpret descriptive statistics
D. Reject the null hypothesis
A. Generalize to a population based on information gathered from a sample
The major factor that affects the standard error of the mean (SEM) is the:
A. Value of the score range
B. Shape of the sampling distribution
C. Sample Size
D. Value of the mean
C. Sample size
For which of the following levels of significance is the risk of making a Type II error greatest?
A. 0.10
B. 0.05
C. 0.01
D. 0.001
D. 0.001
Assuming the normal distribution, about what proportion of a sample lies between the mean & 2 SDs above the mean?
A. 34%
B. 47%
C. 68%
D. 95%
E. None of the above
B. 47%
A statistical procedure that is used to determine whether a significant difference exists between any number of group means is the:
A. t-Test
C. correlation coefficient
D. Mann-Whitney U-test
If a researcher wanted to determine whether observed proportions differed significantly from expected proportions, the statistic would be a(n):
A. t-test
B. correlation coefficient
D. Chi-square
D. Chi-square
The independent variable is weight gain during pregnancy. The dependent variable is the infant's birth weight. The appropriate test statistic is a(n):
A. t-test
C. Chi-square
D. Pearson's r
D. Pearson's r
The independent variable is race/ethnicity (white/nonwhite). The dependent variable is pulse rate. The appropriate statistical test is a:
A. t-test
B. Mann-Whitney U-test
C. Chi-square
D. Pearson's r
A. t-test
The independent variable is marital status (single/married/separated or divorced). The dependent variable is whether or not the person had been hospitalized in the preceding 12 months. The appropriate statistical test is a(n):
A. t-test
C. Chi-square
D. Pearson's r
C. Chi-square
The independent variable is treatment group (visual stimulation/auditory stimulation/no special stimulation). The dependent variable is infants' length of stay in the hospital after birth. The appropriate statistical test is a(n):
A. t-test
C. Chi-square test
D. Pearson's r
As sample size decreases, so does the standard error of mean. T/F
A researcher never knows whether an error has been committed in statistical decision making. T/F
A statistically significant finding means that the obtained results are not likely to have been due to chance. T/F
It would be appropriate for a researcher to test the difference between the means of 3 groups of students by a t-test for independent samples. T/F
Establishing a significance level of 0.01 is more conservative than establishing one of 0.05. T/F
Nonparametric tests have fewer assumptions than parametric tests. T/F
Qualitative research involves an EMERGENT DESIGN . . . a design that . . .
emerges in the field as the study unfolds.

Although Qualitative Design is FLEXIBLE, Qualitative Researches plan for broad contingencies that pose decision opportunities for study design in the field.

Researchers tend to be creative and intuitive, putting together an array of data drawn from many sources to arrive at a holistic understanding of PHENOMENON.

Research question is BROAD.

Individual's perspective is VERY IMPORTANT.

Data collected through interviews or observations.

Concerned with in-depth description of people or events.

Strives for understanding of the WHOLE.

Requires the RESEARCHER to be the INSTRUMENT.
One disadvantage to Qualitative Research
Researcher can que patients for answers, can be BIASED.

Qualitative research forces researcher to define their role & identify their own biases.
When does researcher stop in a Qualitative research project
When the same story is produced over & over. This is when the information becomes SATURATED.
List phases of Qualitative Research Design
1. Orientation & overview - must decide what they don't know & how to handle the phenonmenon

2. Focused exploration - Focused scrutiny & in-depth exploration of those aspects of the phenomenon that are judged to be salient

3. Confirmation & closure - Try to establish that their findings are trustworthy
List common goal of Qualitative Research
To raise consciousness, politicize or activism.

Research focus is on OPPRESSED GROUPS

Study of cultures/behaviors
List only focus of Qualitative Research
to DESCRIBE . . . .is atheoretical

THEORY is developed that is GROUNDED in the data collected
List main Design features of Qualitative Research

May look back to find antecedent factors leading up to the occurrence of that phenomenon

No manipulation or control

Group comparisons not planned but may occur

Can be cross-sectional or longitudinal

Setting is in REAL WORLD

Counts the number of times something occurs

Data coded to identify recurrent themes

Used to describe attitudes, expectations & perceptions

Can be used alone or in conjunction with other methods
List purpose of Qualitative Research
To answer questions concerning causes, effects or trends relating to past events that may shed light on present behaviors or practices
Define Credibility
the "truth" of the findings from the subjects

i.e. "Is this what you were saying?"
Define Transferability
the study's ability to preserve meanings, interpretations & inferences when applied to another similar context
Define Confirmability
obtaining direct & repeated affirmation of what the research has heard, seen or experienced
When is Qualitative Research a good choice?
When a topic has not been studied

When quantitative approach does not or would not give a full picture
Define Grounded Theory
it aims to discover theoretical precepts grounded in data

Researchers try to account for people's actions by focusing on the main concern that the behavior is designed to resolve
Define Systematic Review
"Concise summaries of the best available evidence that address sharply defined clinical questions"

-Cochran Collaboration, 1999

Scientific way to summarize research evidence

Distinguishes interventions that work from those that do not work

Gives relative estimates about how well various options work

Identifies GAPS in KNOWLEDGE
Systematic Reviews of quantitative studies often involve statistical integration of findings through META-ANALYSIS, a procedure whose advantages include:

Enhanced Power

Meta-Analysis is not appropriate for:
Broad questions or when there is substantial inconsistency of findings.
When is it appropriate to use Meta-Analysis?
Research questions/hypotheses in studies are similar

Does not have to be Identical

So Independent & dependent variables & the study population are similar so they can be integrated

Needs to be a sufficient base knowledge for statistical integration

Consistency of the evidence
What is a confidence interval?
Range in which the true effect lies within a given degree of certainty

Usually 96-99%

Confidence that 95% population mean lies between the values equality to + or - standard deviations above & below the sample mean.

Small SD, results are close around the mean

Greater the width of CI, the more dispersed the results
List disadvantages to Systematic Reviews
Potential to combine studies that greatly vary in overall quality

Researcher may fail to fully consider variation in subjects, methods, interventions or outcome measurement
What is the "Ideal number" for NNT (Number Needed to Treat)?
Ideal is 1 Formula 1/ARR
100 is very bad
List purpose to Systematic Reviews
To indicate what we know about the clinical effectiveness of a particular health care process.

Identify gaps in what is known pointing to a need for further research.
List steps to Systematic Reviews
Formulate the research question

Construct an Analytic Framework

Conduct a comprehensive search for evidence most important & time-consuming

Critically appraise the evidence

Synthesize the body of evidence (Meta-Analysis)
T/F Constant Comparison is a technique used in grounded theory in developing theoretical categories.
T/F Case study data can be qualitative or quantitative.
T/F Most qualitative observational data are recorded on videotapes.
The research design for a quantitative study involves decisions with regard to all of the following EXCEPT:
A. Whether there will be an intervention
B. What types of comparisons will be made
C. Whether there will be a theoretical context
D. How many times data will be collected
C. Whether there will be a theoretical context
T/F A Type I Error refers to the researcher's conclusion that no difference exists when, in fact, a difference does exist.
T/F The standard deviation is a measure that tells how spread out scores are in a distribution.
In a correlational study, as compared with an experimental study, the research forfeits control of:
A. The extraneous variables
B. The dependent variable
C. The criterion variable
D. The independent variable
D. The independent variable
T/F A researcher never knows whether an error has been committed in statistical decision making.
This is why research can not be "proven".
T/F A +0.50 correlation coefficient indicates a stronger relation than does a correlation of -0.75.
In qualitative research report, the thematic analysis of the data would be presented in the:
A. Discussion section
B. Methods section
C. Results section
D. Introduction
C. Results section