C. Research Flashcards
________: Prediction of a relationship; Often expressed as more or less than
Hypothesis
________: No relationship; Often expressed as equal to
Null hypothesis
Specific lab, clinical, objective or subjective findings are noted in the ______ section. Results are INTERPRETED in the ______ section.
Results, Discussion
________ research:
Generates hypothesis.
Establishes associations among actors but does NOT determine causal relationships.
Descriptive research
_________ research:
Tests hypothesis.
ALLOWS causal relationships to be determined.
Analytical research
The ________ group receives the treatment, while the _____ group does not, but may receive placebo which gives the aura of treatment.
Experimental, control
________ ______: positive response due to attention received. (AKA placebo effect)
Hawthorne effect
*A Quasi-experimental design is the _____ _____ method. Measurements are taken before and after the program to see whether there has been a noteworthy change.
Time series method
Types of Descriptive Research include:
Qualitative research
Case study
Survey
Types of Analytical Research include:
Experimental model Quasi-experimental design Cohort studies Case-control studies Cross-sectional studies
Uses randomization to divide population into experimental and control groups. Program is successful if experimental group has greater improvement than the control.
Experimental model
Focuses on a specific disease, comparing those with the disease to those without. Both groups recall past behaviors. Retrospective.
Case-control studies
Cohort of healthy people followed through time to see if they develop a specific disease. Sometimes called ‘incidence studies’. Longitudinal and prospective.
Cohort studies
One time data collection. Sometimes called ‘prevalence studies’.
*A snap-shot look at one point in time; describes current, not past or future events.
Cross-sectional studies
Series of measurements at periodic intervals before, throughout, and after the program. Shows whether a noteworthy change has been achieved. (time series method)
Quasi-experimental design
Relevance or ______: the ability to measure a phenomenon it intends to measure
Validity
- ______ ______: Tests whether the difference between two groups is real
Ie. Has the experimental group really performed differently?
Internal validity
- ______ ______: Tests whether a generalization made from the study may apply to a larger population
External validity
________ of ______ (ANOVA) is a tool used to evaluate validity of a test.
Analysis of variance
______: Evaluates consistency or reproducibility of results. _______ is a measure of this.
Reliability, precision
______: Measures the proportion of actual positives which are correctly identified
______: Measures the proportion of negatives which are correctly identified as such
Sensitivity
Specificity
______ variables are outcomes. _____ variables are what is manipulated in the study. Treatments for diseases are ______ variables.
Dependent, Independent, Independent
Sampling:
__________: each segment of the population will be represented in the sample
(i.e. randomization)
__________: No way of forecasting that each element will be represented in the sample
(i.e. convenience or accidental - no attempt to control bias; quota - select units in same ratio as found in general population)
Probability
Non-probability
Measuring Data Distributions
Simple average
Mean
Measuring Data Distributions
Midpoint
Median
Measuring Data Distributions
Most frequently occurring value
Mode
Measures of Dispersion
Difference between highest and lowest value in the set
Range
Measures of Dispersion
Indicates degree of dispersion about the mean
Standard deviation
*2/3 (68%) of all observations in a normal distribution lie within _ Standard Deviation of the mean.
95% lie within 2 SDs of mean.
*2/3 (68%) of all observations in a normal distribution lie within 1 Standard Deviation of the mean.
Existence of relationships
*The ______ the points are to the line, the ______ the degree of linear relationship. (on a scatter diagram)
The closer the points are to the line, the stronger the degree of linear relationship.
Linear correlation coefficient (r).
The value of R is always between __ and __.
The value of R is always between -1 and +1.
The ______ r is to 1 OR -1, the ____ the degree of the linear relationship.
If r=0, there is no linear relationship.
Closer, stronger
Level of Significance
The _____ the p-value, the ____ the significance of your results.
Lower p-value, Higher significance
(most will not accept results as statistically significant unless p
This type of study is used to remove bias from research; neither the researcher nor the subject knows which group is receiving treatment and which the placebo
Double blind study
Rate of death
Mortality
State of disease
Morbidity
Techniques that allow conclusions to extend beyond an immediate data set; Infers qualities about a population; To make generalizations from a sample to a population.
Inferential statistics
This test does not depend on a normal distribution
Nonparametric test
Only 2 events are possible in this score (i.e. heads or tails)
Dichotomous score
Scores measured on a continuous scale
Continuous score
This type of study is a scaled down version of the larger investigation. It includes every step in the study, but done on a small test group to test for feasibility.
Pilot study
Method of obtaining information about a target group through this smaller group who discuss beliefs, opinions, problems, etc.
Focus group
This test whether or not there is a real difference between categories. Compares the frequency with which observations are expected to occur with the frequency they actually occurred.
Chi square
Tests null against alternative hypothesis.
If this probability value is less than or equal to the level set for significance, the null hypothesis is rejected, in favor of the alternative hypothesis.
T test
