Part 2 Flashcards
What does Quantitative Methodology emanate?
Positivist/post-positivist perspective
What is the focus of Quantitative Methodology?
predominant ‘biomedical’ focus
What does Quantitative Methodology use objectivity, logic, experimental/scientific processes to do?
- compare
- infer causation
- predict outcomes
What is Descriptive Quantitative research?
structured observations or survey (or both) used to describe a phenomenon, situation, group or characteristic
What is Exploratory Quantitative research?
to gain new insights, discover new ideas & increase knowledge about a phenomenon (usually conducted when little is known about a topic)
What is Causal Quantitative research?
experimenting (statistically speaking) to assess cause and effect
What is the purpose of a research design?
- answer research question
- control for/understand biases
- direct analysis
- direct interpretation
What does the design of a research project do?
- directs the research process
- framework of who, what, when, where & how
What is a Variable?
a measurable characteristic that varies among subjects
What are the different types of variables?
- independent
- dependent
- extraneous variable or covariate
What is an Independent Variable?
interventions or presumed cause
ex. salt intake
ex. self-care symptom management
What is a Dependent Variable?
outcome or presumed effect
ex. blood pressure reading
ex. quaility of life
What is an Extraneous Variable or Covariate?
alternative or other possible causes
ex. Sedentary
ex. severity of illness
What is Control?
the measures that the researcher uses to hold the conditions of the study uniform and avoid possibe impingement of bias on the dependent variable or outcome
What are the different ways to control for Extraneous Variables? (6)
- utilizing a homogenous sample
- random assignment to groups
- consistent data-collection procedure
- minimize threats to internal validitiy
- using experimental design (manipulation of independent variable)
- statistical manipulation
What is a Natural Setting?
uncontrolled, real life situation
What is Partially Controlled?
manipulated or modified in some way
What is Highly controlled?
Artifical environment for sole purpose of doing research
What is Internal Validity?
the extent to which the observed results represent the truth in the population we are studying and, thus, are not due to methodological errors.
What is External Validity?
the extent to which the results of a study are generalizable to patients in our daily practice, especially for the population that the sample is thought to represent
If a study shows a high degree of internal validity, what does that mean?
We can conclude we have stronge evidence of causaility
If a study shows a high degree of External validity, what does that mean?
we can conclude we have little or no evidence of causality
What are threats to Internal Validity? (6)
- selection bias
- instrumentation
- history
- maturation
- testing
- mortality
What do threats to external validity do?
compromise our confidence in stating whether the study’s results are applicable to other groups
What are the threats to external validity?
- selection effects
- reactivity
- measurement effects
What study designs control threats to validity?
- Experimental: control most
- Quasi-experimental
- Non-experimental
- Descriptive
What are the Cause-effect classes?
- experimental
- Quasi-experimental
What are the Description classes?
- non-experimental
- descriptive
What is an Experimental Design?
- there is an intervention that is controlled or delivered
- there is an experimental and control group
- there is a random assignment to groups
What is Random Assignment to groups?
- internal validity issue
- equal chance that extraneous variables exist in both groups
- experimental/control group assignment
What is Random Selection from population to sample?
- external validity issue
- equals extraneous variables in the sample that are true for the population
- sampling strategy
What are the strengths of Experimental Design?
- establish causality/causal direction
- control
What are the limitations of Experimental Design?
- may be difficult to implement
- generalizability (external validity) may be low
- not ethical for some conditions
What is a Quasi-Experimental Design?
- there is an intervention that is controlled or delivered
- there is an experimental group with or without a control group
- there is typically no random assignment to groups
What are the Strengths of Quasi-Experimental Design?
- establish causality/causal direction
- control
- practical, real world adaptability
- people not always willing to be randomized, so this works for them
What are the limitations of Quasi-Experimental Design?
- cannot make clear cause-and-effect statements
- generalizability (external validity) may be low
- not ethical for some conditions
What is Non-experimental Research?
research that lacks the manipulation of an independent variable
What are the two types of non-experimental research?
- Survey Studies
- relationship/difference studies
What are Survey Studies?
- structured interview
- self-report: Questionnaires, online surveys/computer-assisted
- usually a large number of respondents
What is the purpose of Survey Studies?
to obtain info about the prevalence, distribution, and interrelations of phenomena within a population
What are the advantages of Survey Studies?
- flexible & broad in scope
- data can be collected in a number of ways
What are the disadvantages of Survey Studies?
- data relatively superficial, rarely probes deeply into human complexities
- low response rates vs. number of participants needed
- limitations with self-report
What is the purpose of Relationship/Difference Studies?
to look at the relationships or differences between variables that can provide a deeper insight into a phenomenon
What are the different Relationship/Difference Studies?
- correlational
- developmental
What are the different types of Development Relationship/Differences Studies?
- cross-sectional
- longitudinal/prospective
- retrospective/ex post facto
What are Correlation studies?
- involves the measurement & determination of the relationship between 2 variables
- cannot be used to determine cause-and-effect
What is Postive Correlation?
one variable increases, scores on the other variable also increases (r values ranges from 0 to +1)
What is Negative Correlation?
increase in one variable is accompanied by a decrease in the second variable (r values ranges from -1 to 0)
What is the Correlation Coefficient (r)?
numerical index that reflects the relationship between two variables. Value: between -1 and +1
What are the advantages of Correlational Studies?
- can be applied to many populations, focus on a wide range of topics, & info can be used for many purposes
- a framework for exploring the relationship between variables that cannot be inherently manipulated
What are the disadvantages of Correlation studies?
- correlation does not equal causation
- no randomization in the sampling procedures because the study deals with pre-existing groups; therefore, generalizability is decreased
What is Cross sectional Research?
examine the relationship between variables of interest as they exist in a defined population at one particular time
What are the advantages of Cross-sectional Research?
- fast
- less expensive
- usually involve large number of participants
- can collect a large number of variables
What are the disadvantages of Cross-sectional Research?
- no causal association
- harder to infer changes over time
- threat to internal/external validity
What is Prospective (longitudinal cohort) Research?
one or more groups are followed across time and compared on one or more variables
What are the Strengths of Prospective Research?
- time line established
- large number of participants
- large number of variables
What are the limitations of Prospective Research?
- very expensive, long term follow-up needed
- large loss to follow-up possible
- threat to validity
What is Retrospective Design Research?
an effect (outcome or DV) observed in the present is linked to a potential cause that occured in the past
What are the strengths of Retrospective Research?
- fewer participants
- large number of variables
What are the limitations of Retrospective Research?
- very difficult to find adequate control group
- beware of alternative hypothesis as the reason for documented relationship
- threats to validity
What is an Accessible population?
aggregate of cases that conform to designated (eligibility) criteria and are accessible for a study
What is Sampling?
Process of selecting subjects (elements) to represent a population
What is Nonprobability Sampling?
does not involve selection of elements at random; is rarely representative of the population
What is Probability Sampling?
- involves random selection of elements
- allows researchers to estimate the magnitude of sampling error
What is important to remeber about Random probability?
- Random probability does not equal randomization of participants
- randomization increases internal validity
- random sampling increases external validity
What are the types of Probability sampling?
- simple random
- stratified random
- cluster
- systematic
What is Simple Random Sampling (SRS)?
- needs a sampling frame
- each element has an equal & independent probabillity of selection
- uses a random number generator, random number table etc.
What is Systematic Sampling?
- needs sampling frame
- uses the frame’s order to locate element
- faster than simple but may introduce bias IF the frame is arranged in a way that coincides with sampling occurence
What is Stratified Random Sampling?
- the population is first divided into two or more strata, from which elements are randomly selected
- aim is to enhance representativeness
What is Disproportional Sampling?
- a probability sampling technique used to address the difficulty researchers encounter with stratified samples of unequal sizes
- requires “adjustments” or “weighting” to arrive at estimates of overall population values
What is Multi-stage Cluster Sampling?
- involves selecting broad groups (clusters) rather than selecting individuals, and is typically the first stage of a multistage approach
- less accurate than simple or stratified random sampling
- BUT more practical than other types of proability sampling
What are the different types of Non-probability Sampling?
- convenience
- quota
- purposive
- network/snowball
- consecutive
What is Convenience Sampling?
- use elements available at time & place of study
- does not require sampling frame
- can be fast and efficient but may not provide representative sample - people who are readily available might be atypical of the population
What are Quota Sampling?
You first divide the population into strata and then recruit sample units until you reach your quota (predetermined number)
What is Consecutive sampling?
recruiting all people from an accessible population over a specific time interval
What is Purposive Sampling?
uses researchers’ knowledge about a population to hand sample members
What is Heterogeneous Sample?
participants are diverse
What is Homogeneous Sample?
- participants are similar
- generalizability may be limited to people who are similar to your target population
Describe the Sample Size:
- estimated before the study to determine number of elements required to demonstrate treatment effect
- a large sample size does not make up for non-representative sample
What occurs if the sample size is too small?
can result in type II error or low “power”: non-significant result due to too few observations rather than because intervention is not effective/no relationship between IV/DV
What occurs if the sample size is too large?
unnecessary cost without added benefit to study results
What is Power?
the likelihood that a researcher will find a significant result (an effect) in a sample if such an effect exists in the population being studied
- if there is not enough power in a study, it would not matter how big the effect size
- commonly set a 0.80 (1-beta)
What is Power Analysis?
procedure used to estimate sample size requirements prior to undertaking a study (to diminish the likelihood of type II error)
- not usually required for pilot studies
What is Alpha?
probability of making a type I error (commonly set at 0.05 or 0.01)
What is Beta?
probability of making a type II error (usually 0.2 or 20%)
What is Effect Size?
measure of the magnitude of the effect (group difference) is
Describe a Small Effect Size:
- correlations around 0.20
- requires larger sample size
Describe Medium Effect Size:
- correlations around 0.40
- requires medium sample size
Describe Large Effect Size:
- correlations around 0.60
- requires smaller sample size
What are the different Data Collection Methods?
- biological or physiological
- observational
- Questionnaries, surveys/instruments
- records or available data
- interviews
What are the different types of Biological/Physiological Measures?
- physical
- anatomical
- chemical
- mircobiological
What are the advantages of Biological/Physiological Measures?
- objective
- precise
- sensitive
What are the disadvantages of Biological/Physiological Measures? (5)
- can be invasive
- expensive
- hard to obtain
- may need special training
- may cause reactive effects
What are the different informations that Observational Methods can obtain? (5)
- patient’s conditions
- verbal communication
- non-verbal communication
- activities
- environmental conditions
What are the advantages of Observation methods?
ideal for studying complex interactions & measuring people’s actions/reactions
What are the disadvantages of Observation methods?
- reactivity effects
- observer bias
What are the advantages of Records?
usually inexpensive, historical
What are the disadvantages of Records?
- availability
- ethics
- bias
- missing data
What are the advantages of Questionnaries?
- fast
- economical
- variety
- participants can remain anonymous
What are the disadvantages of Questionnaries? (5)
- breadth vs. depth
- response rates
- recall bias
- social desirability bias
- incomplete items
What is Social Desirability Response Bias?
giving answers that are congruent with prevailing social values
What are Extreme Responses?
selection of extreme alternatives
What is Internal Reliability?
Stability within an instrument - looking to ensure items measure the same attribute/construct
How is Internal Reliability measured?
Cronbach Alpha statis - normal range is between 0 to +1, higher values are associated with greater internal reliability
What is the Test-retest correlation coefficient?
stability over time - ability to obtain similar scores in different situations/repeated testing
What is the Reliability Coefficient?
is a way of confirming how accurate a test by giving it to the same subject more than once and determining if there’s a correlation and similarity between the two scores
- over 0.80 as a result is good
What is Interrater Reliability?
the degree of agreement among independent observers who rate, code, or assess the same phenomenon
What are the 3 types of Content Validity?
- Content
- criterion-related
- construct
What is Content Validity?
evaluates if the test is fully representative of what it aims to measure
What is Criterion-related Validity?
measures if the results accurately measure the outcome they are designed to measure
What is Concurrent Validity?
shows you the extent of the agreement between two measures taken at the same time
- subtype of criterion-related
What is Predictive Validity?
ability to predict future behaviour/future correlation with a measure of the same concept
What is Construct Validity?
the degree to which the test measures the concept that it’s intended to measure
What is Convergent Validity?
how closely a test is related to other tests that measure the same thing
What is Divergent Validity?
Shows that concepts or measurements that might be in the study is different/unrelated to what you are actually studying
Describe the difference between Validity and Reliability?
- validity is about the appropriateness of a test
- reliability is about the consistency of the scores produced
What is Capture Variability (Variance)?
how the scores vary across participants and identify what is typical and atypical among the data
What is Data Reduction?
summarizing or describing the characterisitcs of a set of data
What is the Mean?
the average value from a data set
What is Standard Deviation (SD)?
- the most common measure of variance
- describes how far the values stray from the mean
What is Inferential Statisitcs?
- allows you to make predictions (“inferences”) from that data
- you take data from samples and make generalizations about a population
What is the P-value?
the probability that the results were due to chance and not based on your program/intervention
What occurs if the P-value is low?
the lower the p-value, the most likely it is that a difference occurred as a result of your program
What occurs if the P-value is less than the alpha value?
you can conclude that the difference you observed is statistically significant
What is the Confidence Interval?
a range around a measurement that conveys how precise the measurement is
What is the long-held minimum threshold for acceptance?
95% or p < 0.05, meaning that the threshold is considered statistically significant
What does Statistically Significant mean?
indicates that the results were unlikely to be due to chance
What does the absence of statistically significant mean?
it does not mean that the finding is unimportant
What is Mixed Methods Research?
a procedure for collecting, analyzing and “mixing” both quatitative and qualitative research methods in a single study to understand a research problem
What is Inductive Reasoning?
- details to general picture
- foundation of most qualitative research
- theory generating
What is Deductive Reasoning?
- general picture to specific explanantions
- foundation of most quantitative research
- theory testing
What are the different types of Mixed Methods Designs?
- Convergent Parallel Design
- Explanatory Sequential Design
- Exploratory Sequential Design
