Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
91 Cards in this Set
- Front
- Back
External validity
|
the extent to which the results of a study can be generalized.
|
|
selection bias
|
the sampling procedures favour the selection of some individuals over others.
|
|
a biased sample
|
has different characteristics from those of the population
|
|
selection bias
|
the sampling procedures favour the selection of some individuals over others.
|
|
target population
|
complete group of individuals of interest to the researcher.
|
|
accessible population
|
portion of the target population who are accessible to the researcher.
|
|
sample
|
individuals who are chosen to participate in the study ( intended to represent population)
|
|
population
|
all of the individuals of interest to the researcher.
|
|
Probability Sampling
|
each member of the population has a specifiable probability of being chosen to participate.
Important for applied research and when you wish to characterize a population. Most probability sampling techniques require a sampling frame which is a complete list of all of the members of a population |
|
simple random sampling
|
each individuals in the sample has an eqal chance of being selected. individuals are randomly selected from a sampling frame.
|
|
Systematic random Smapling
|
one individual is randomly selected from a sampling frame and then others are chosen at regular intervals.
|
|
stratified random sampling
|
the population is divided intro strata and individuals are randomly selected from each stratum.
|
|
proportionate stratified sample
|
random samples are drawn from each stratum in a way that is proportionate to the population
|
|
Disproportionate stratified sample
|
random samples are drawn from each stratum in a way that is disproportionate to the population.
|
|
cluster sampling
|
clusters of individuals are identified, a random sample of the clusters are chosen and all members in those clusters are sampled.
|
|
unbiased sample
|
a complete and accurate sampling frame
a high response rate the process of randomization to work. |
|
purposive sampling
|
sampling with a purpose in mind, to obtain a sample of people who meet some predetermined criterion.
|
|
snowball sampling
|
one or more individuals from the population are contacted and these individuals are used to lead the researcher to other population members.
|
|
Nonprobability sampling
|
the probability of any particular member of the population being chosen to participate is unknown. It is more convenient and inexpensive. Sampling frames are typically not possible to explain. representativeness of the sample is less important.
|
|
Quota sampling
|
people are selected haphazardly, but in a way to obtain specific percentages of people with different characteristics.
|
|
Generalization issues
|
volunteer bias, gender bias, cross-species generalizations.
|
|
operational definitions
|
make concepts public.
Precise statements of how an abstract construct is translated into a measurable variable. |
|
Construct validity
|
the degree to which the measure actually measures the construct it is intended to measure.
|
|
discriminant validity
|
the extent to which scores on two different measures of different constructs do not relate to each other.
|
|
convergent validity
|
the extent to which scores on two different measures of the same construct relate to each other.
|
|
Measured score
|
True score + Random Error
|
|
True score
|
the score you would obtain if there was no error in the measure
|
|
Random Error
|
extraneous factors that influence your score on the measure.
Observer Error Environmental Changes Participant Changes. |
|
Parallel forms reliability
|
the extent to which scores on two different, but equivalent versions of the same measure, administered at two different times, correlate with each other.
|
|
internal consistency
|
the assessment of reliability using responses at only one point in time.
|
|
Conbrach's Alpha
|
the correlations between each item and every other item are calculated and the average of all those correlations is computed.
|
|
Split half reliability
|
the test items are divided in half, scores on each half are calculated and the correlation between the scores on the two halves is computed.
|
|
Reliable measures
|
are not necessarily valid
But VALID MEASURES MUST BE RELIABLE |
|
Modalities of measurement
|
Self report measures - participants provide reports on themselves
Physiological measures - responses of the participants bodes are recorded ( GSR, EEG, fMRI) |
|
Behavioural Measures
|
researchers directly observe participants' behaviour
|
|
Nominal scales
|
involve meaningful but potentially arbitrary and non-numerical names or categories
|
|
ordinal
|
involve ordering or ranking the levels of the variable
|
|
interval scales
|
involve the use of real numbers designating real amounts that reflect relative differences in magnitude
|
|
ratio scales
|
contain an absolute zero point that represents the absence of the variable being measured.
|
|
Sensitivity of measures
|
Floor Effect - participants all perform at the bottom end of the measurement scale.
Ceiling effect - participants all perform at the top end of the measurement scale. Continuous measures are more sensitive than binary measures. |
|
Tri-council policy statement
1998 |
tree main canadian funding agencies jointly published the TCPS .
Medical research council ( Canadian institutes of health) Social Sciences and Humanities research council Natural Sciences and Engineering Research council Any institution that receives funds from one of these agencies must adhere to the TCPS |
|
Tuskegee syphilis study
|
physicians infected 400 low incme, mostly illiterate black men with syphilis to study the course of the disease. treatment was refused even after ac ure was discovered in 1943.
Willowbrook studies- physicians infected about 800 mentally impaired children with hepatities to study the course of the disease. Nuremberg Code (1949) - set of 10 guidelines for the ethical treatment of human participants in research |
|
TCPS Overview
|
Section 1 : Ethics Review
Section 2: Free and Informed Consent Section 3: Privacy and Confidentiality Section 4: Conflict of Interest Section 5 : Inclusion in Research. |
|
REB ( research ethics board's structure )
|
the reb membership should be broad enough to reflect the community or society from which research subjects are recruited. 5 people. 2 in areas of research. one in ethics. one has to come from community that REB services. Minimal risk is reviewed by one - anything more is 5
|
|
differential design
|
demonstrates the existence of a relationship between two variables.
|
|
Ethics Review (Section 1)
|
harms and benefits, minimal risk
|
|
Free and Informed Consent (Section 2)
|
Informed Consent, Coercion, Competence. Deception. Debriefing
|
|
Informed consent is not necessary if
|
the research involves no more than minimal risk.
- lack of consent is unlikely to adversely affect the rights and welfare of the participant - the research can not practicably be carried out with informed consent. - there is no therapeutic intervention and whenever possible and appropriate the participant is debriefed |
|
Privacy and Confidentiality (section 3)
|
Limits, Anonymity, Privacy and confidentiality
|
|
Conflicts of interest (section 4)
|
Involving researchers,
involving the institution involving REB memebrs. |
|
Inclusion in Research (Section 5)
|
Distribution of benefits
|
|
professional codes of ethics
|
canadian code of ethics for psychologists. APA ethical guidelines.
Canadian council on animal care's code of ethics. |
|
Method of tenacity
|
accepting ides as true because they have been believed for a long time or because of superstition
|
|
method of intuition
|
accepting ideas as true basedo n hunches and gut feelings
|
|
method of authority
|
accepting ides as true because some expert or authority figure asserts them to be true.
|
|
rational method
|
involves developing ideas using existing ideas and the principles of logic
|
|
empirical method
|
observation and experience are used to obtain knowledge .
|
|
Theory
|
inductive reasoning ( specific - general )
an interrelated set of concepts used to explain ab ody of data and make new predictions about the results of future studies. |
|
Research Hypothesis
|
a specific prediction about what will happen in a particular study.
Deductive reasoning ( general -- > specific ) |
|
( Step 1 )Scientific Method and Research process.
|
Identify a Problem of Issue
|
|
Step 2 Scientific Method and Research process.
|
Narrow down the issue
A) identify variables to measure B) define variables |
|
Step 3 Scientific Method and Research process.
|
Develop a research hypothesis
a) logical b) testable c) refutable/falsifiable d) positive |
|
Step 4 Scientific Method and Research process.
|
Design and conduct the study
a) identify participants b) select a research strategy ( experimental vs non experimental) c) select a research deisgn d) obtain ethical clearance e) conduct the study. |
|
Step 5 Scientific Method and Research process.
|
Analyze the data and interpret the results
support refute or refine the hypothesis |
|
Step 6 Scientific Method and Research process.
|
Comunicate the results
|
|
Elements of science
|
Science is empirical ( systematic empiricism is used so that observations are carefully structured and controlled)
Science is public peer review , replication Science is objective |
|
Components of a good theory
|
general
parsimonious - they provide the simplest possible amount of the outcomes conducive to future research falsifiable - they can be shown to be wrong. |
|
Correlational research strategy
|
Goal - to describe the strength and direction of relationship between variables
|
|
Correlational strategy
|
two or mroe variables are measured as they naturally occur and the extent to which they go together or correlate is examined.
|
|
Shapes of relationships
|
Linear, Curvilinear, No relationship
|
|
Nonmonotic
|
the relationship between two variables changes direciton at some point
|
|
monotonic
|
the relationship between two variables remain in the same direction
|
|
Phi coefficient
|
both variables are nominal and dichotomous.
|
|
point biserial coefficient
|
one of the variables is nominal and dichotomous and the other is continuous.
|
|
pearson product moment correlation coefficient
|
both variables are on an interval or ratio scale.
|
|
Correlation does not equal causation
|
third variable problem, directionality.
|
|
Regression
|
involves using information about a relationship between variables to make predictions of one variable on the basis of another variable
|
|
Multiple regression
|
involves predicting one variable on the basis of more than one other variables
|
|
Partial correlation
|
a statistical technique used to control for third variables.
|
|
Clinical Equipoise
|
clinicians have an ethical responsibility to provide the best possible treatment for their patients.
|
|
Belmont report
|
provides a common set of federal regulations for protecting human participants to be used by review board.
|
|
criterion variable
|
the variable we predict ( ordinal, ratio, interval ) can only predict linear
|
|
predictor variable
|
the variable we use to predict
|
|
Multiple Regression
|
big R and R2
|
|
correlation
|
correlation of 0 does not mean absence negative correlation does not rule positive causal relationship
|
|
third variable
|
spurious (false) correlation
|
|
correlation
|
small correlations based on a large sample is statistically significant.
|
|
test reliability retest must be .80 to be acceptable.
|
x.
|
|
correlation
|
cannot be used if on nominal scale and more than 3 categories.
|
|
correlation relationship
|
must be monotonic and linear. if not will underestimate the relationship between two variables. Wise to create scatterplot before correlation.
|
|
correlation
|
direction of relationship,
form, consistency |