before midterm basics Flashcards
define descriptive research
describes characteristics of existing phenomena (at a given time and place)
- provides a broad picture
- serves as a basis for other types of research
- does not include treatment or control groups
does not imply causation
describe historical research
describes past events in the context of other past or current events
explain correlational research
asks what several events have in common
- asks whether knowing one event is assoicated with another event
does not imply causation
what is non-experimental research and what are the methods used?
describes a relationship between variables
Methods:
- descriptive
- historical
- correlational
- qualitative
cannot test cause and effect relationships
describe qualitative research
examine behaviour in a natural social, cultural and political context
usually results in non-qualitiative data
what is experimental research and what are the 2 types?
tries to discover casual relationship
types:
true experimental
quasi experimental
describe true experimental
participants are assigned to groups
the treatment variable is controlled by the researcher
control of potential causes of behaviour
explain quasi experimental
participants are pre-assigned to groups
- useful when researcher cannot manipulate variables
what are variables?
a class of measurements that can take more than one value
there are 2 primary types of variables:
dependent variable & independent variable
what is a dependent variable(DV)?
a variable that is measured to see whether the treatment or manipulation of the independent variable had an effect
- its value depends on changes in the independent variable
the effect
what is an independent variable?
a variable that is manipulated to examine its impact on a dependent variable
- its value is independent of other variables in your study
the cause
so we know the 2 main types of variables but what are the 2 main sub-types?
categorical (discrete)
- fixed number of values (e.x. dead/alive, obese/healthy/underweight)
continuous:
- can have any value between a theoretical minimum and maximum
(e.x. birth weight, BMI, temperaure)
what is a control variable?
a variable that is related to the dependent variable.
- the influence of which needs to be removed
what is an extraneous variable?
a variable that is related to the dependent variable or independent variable that is not part of the experiment
what is a hypothesis?
reflects the general problem under study
- it restates the general problem in a form that is precise enough to allow testing
null hypothesis vs research hypothesis
explain the null hypothesis
states that there is no relationship between the independent and dependent variables under study
what is the purpose of the null? and when is it accepted as true?
it is a starting point for analysis
- accepted when absent of other information
provides a benchmark for comparison
what is a research ypothesis?
a statement of unequality
- a relationship exists between the independent and dependent variables
explain directionality of the hypothesis
non directional hypothesis:
- two tailed
- predicts that the independent variable will have an effect on the dependent variable, but the direction of the effect is not specified
(just states that there will be a difference)
directional hypothesis:
- a prediction made by a researcher regarding a positive or negative change, relationship, or difference between two variables of a population
( this prediction is typically based on past research, accepted theory, extensive experience, or literature on the topic
what is the purpose of an alternative hypothesis?
it is directly tested during the research process to compare agasint the null hypothesis
null vs research hypothesis
null: equality between variables
- refers to population
- indirectly tested
Research: inequality between variables
- refers to a sample
- directly tested
characteristics of a good hypothesis
stated in declaritive form
posits a relationship between variables
reflects on theory or literature
breif and to the point
is testable
results from a sample should _ to the population
generalize
what is the concept of significance?
statistical concept is applied to differences between 2 or more samples
- experiments always assume the null
- statistical significance is used to test differences
how is the concept of significance used in research
the P value: (probability that the null hypothesis is true)
if p is less than .05 then reject the null
if p is greater than .05 then accept the null
explain the flow of research
- idea
- research question
- research hypothesis
- literature review
define general, secondary and primary sources when reviewing literature
General:
- provides an overview of a topic and gives leads to where more info can be found
(newspapers, magazines)
secondary:
- provides a level of information “once removed” from the original work
( books on specific subjects and reviews of research)
Primary:
- the original reports of the orignial work or experience
( journals, abstracts, scholarly books, educational resources)
what are the pros and cons to using journal sites?
pros:
- excellent source
- specific
cons:
- need name of resource
- subscription
what are the pros and cons to using library databases?
pros:
- free through library
- ideal for indepth research on complex topics
- written by professionals
cons:
- can be hard to find simple explanations
- full text isnt always available
pros and cons to using google/ google scholar
pros:
- simple to use
- wide variety of materials
- save to read later
cons:
- wide range but not comprehensive
- doesnt provide notice of when its materials are updated
explain surveys in research
assesses construct such as attitudes, beliefs etc.
can be done with pen and paper method or by interview
multiple types of questions
what types of questions are used in surveys
structured questions:
- possible answers are pre identified
(sex, level of education, year of study)
open-ended questions:
- no single answer
(reason for participation, ideal travel destinations)
using both would be considered a mixed methods design
how to do survey research:
4 steps to the plan
- clarify the objectives
- identify a sample
- define a method
- code and score
explain the pros and cons to the pencil and paper survey method
pros:
- fast, sample size and ananonymous
cons:
- impersonal, no probing, no control over answers, flexibility
what is correlational research?
implies that variables share something in common
does not imply cause and effect relationship
what is a correlation coefficient?
a number between −1 and +1 calculated so as to represent the linear dependence of two variables or sets of data
- (the closer it is to 1 it implies there is alot of shared variability
the strength of relationship is indicated by the absolute value of the coefficient
change in x = change in y
what are the types of correlations?
3
no correlation
positive (direct):
- increase in x = increase in y
negative (inverse):
- increase in x = decrease in y
what is the coefficient of determination?
the squared value of correlational coefficient
- proportion of variance in one variable explained by variance in the other
what is the coefficient of alienation
1 - coefficient of determination
- proportion of variance in one variable unexplained by variance in the other
what are the roles of quasi experimental designs?
investigates differences between groups (could also look at relationships)
- can control extraneous variables
- can compare variables of interest (statistically)
what are the characteristics of quasi experimental designs?
compares groups that are pre-existent
- ethical issues
explain the difference between random selection and random assignment
random selection:
- randomly selecting members of a population for study sample
random assignment:
- sorting sample into control and experimental groups (group 1 and group 2)
- helps to ensure groups are comparable
what happens if we dont randomly select or assign?
sample will likely be biased in some way & data will not represent the entire population
what are the 4 certain types of quasi designs?
one group post test
one group pre/post test
two groups pre/post test
non-equivalent control
one group post test
only testing one group after the experiment
one group pre/post test
testing one group before and after experiment
referred to as ABA or OXO
two group pre/post test
both pre-determined groups get questionnaire before and after test however only one group is given the intervention
ABAB or OXOX
2 group post test
pre-determined groups
one group is the control group
provide treatment to experimental group and assess both post test
non-equivalent controls
predetermined groups examined
- the same as 2 group pre/post test but the groups differ and they both recieve the same treatment
ex. group 1 = adults living with elderly parents
group 2 = parents living alone
- questionnaire is given to both groups before
- both groups partake in same treatment
- post test is given to see results
treatment is manipulated
conclusion about pre-post tests and their cofounding variables
maturation:
- individuals change over time
- (follow up)
practice effects:
find what is influencing the scores
(group 1 recieves 1 treatment, group 2 recieves another treatment. post test shows that both groups increase on the post test)
- (first time didnt know what to expect, second time you have an idea)
mulitple treatments:
- individual participates in multiple sessions
(session 1: memorize words with the music on)
(session 2: memorize words without music)
regression to the mean:
- pre-selecting groups based on performance on. a chance event
- (those who over/under perform will go back to their chance level (mean)
- (ask once then give a 2nd test using the same stratgey, the mean score would be expected to be close to 50)
what are the other 4 popular designs of quasi experimental?
meta-analysis
longitudinal
cross sectional
sequential
describe meta-analysis
give pros and cons
not an experiment
- a statistical process that combines the data of multiple studies to find common results and to identify overall trends.
Pros:
- powerful method to review literature
- allows researcher flexibility
- data collection is easier bc you dont have to deal with people
cons:
- time consuming
- complex method
- box-score method is not as useful in todays society
box score method and its cons
meta-analysis
yes or no questionnaire type thing
- a way of summarizing the positive responses from a Likert scale survey question. It combines the highest 2 responses of the scale to create a single number
cons:wider margins of error
need a larger sample size to detect differences
explain systematic reviewing
meta’s close cousin
reviews literature
- focuses on checklist for quality to research which is used to inform conclusions
looks at studies as a whole
longitudinal design
give pros and cons
individuals are followed over time (observational)
- multiple time points are taken
- allows identification of trends and causation ( if the cofounds are well controlled)
pros:
- cause and effect conclusions
- high comparability across time
- high detail in study design/findings
cons:
- time to execute is very long
- expensive
- high drop out rates
cross-sectional designs
give pros and cons
- snapshot in time of a phenomena (looking at the present moment)
looking at individuals of difference ages at one point ( not following them over time)
(example: comparing individuals ages 25, 30, 35, and 40 at one time) - does not lead to cause and effect conclusions because of inherent variability and preassigned groups
pros:
- inexpensive
- quick
- low drop out rates
cons:
- limits comparison between groups
- no information about participant development
- same age does not equal same development level
- cross generational conclusions can be problematic (cohort effects)
living in different worlds (cohorts)
sequential designs
combines both longitudinal and cross sectional into one study
- assesses individuals ages 25,30 & 35 over time
cons: still going to have cohort effects
pro: will tell you more info than just doing a cross-sectional or longitudinal
retrospective design and what types of questions you might see in this type of interview
quasi design
groups are pre-selected
looks at lifestyle/past events of people
can also focus on cross sections of people
does not allow for cause and effect conclusions
flexibility to assess both qualitative and quanititative:
questions:
qualitative: why did you play that sport at age 10-20
quantitative (categorical): highest education level
quantitative (continuous): how many hours a week did you play sport at ages 10-15
