Experimental design Flashcards
What is accuracy?
How close a measurement is to the true value of the quantity being measured.
What is precision?
How closely a set of measurement values agree with each other but gives no indication of how close the measurements are to the true value.
What are systematic errors due to?
- Environmental factors
- Observational/researcher error
- Incorrect measurement instrumental calibration
Hypothesis
It is predicted that (participant/population) who/when (IV experimental condition), will have/show (direction) (DV) compared to those who/when (IV control condition)
What is the ethical concept integrity?
- Honesty
- Reporting ALL findings
Objective and open reporting and recording of results, processes of peer review. Thoroughness of any literature review and other research procedures.
What is external validity?
- Only considered when internal validity is present
- Sample should show population
- Study= real life setting to some extent
What are the 2 types of validity?
Internal= extent to which an investigation measured or investigates what it claims to
External= only considers when internal validity is present
What is validity?
Extent to which psychological tools and investigators truly support their findings or conclusions. A valid measure is on that measures what it intends to measure.
What is reproducibility?
The extent to which successive measurments or studies produce the same results when repeated under different conditions.
What is repeatability?
Some researchers repeat their study using all the same procedures.
Uncertainty in data?
Refers to the lack of exact knowledge relating to something being measured due to potential sources of variation of knowledge.
Fieldwork- Yarning circles
In aboriginal and Torres Strait Islander cultures, this is a traditional approach to group discussion which involves talking, exchanging idea, reflection and deep considered listening without judgment.
Fieldwork- Direct observation
A researcher watches and listens to the participants of a study, with no direct intervention and involvement, or manipulation of variables.
Fieldwork- Questionnaires
A set of questions or prompts given to participants to answer digitally or with pen and paper. Questions may be open-ended or closed.
Fieldwork- focus groups
A qualitative research method which involves a researcher conducting a discussion with a small group of people on specific topics.
Fieldwork- Qualitative interview
Involve a researcher asking questions to gather in-depth information about a particular topic, theme or idea. Generally open ended questions.
- Provies rich, qualitative date
Random errors may be reduced by…
- Repeating and conducting more measurments
- Calibrating measurment tools correclty
- Refining measurment procedures
What are random errors due to?
- Chance
- Unsystematic
- Random
What does the ethical concept of respect mean?
The consideration of the extent to which living things have an intrisnic value and/or instrumental value.
What does Non-malificence mean?
- Does it do harm?
- Avoid harm
What is the ethical concept benificence?
- Minimising risks and maximising benifits
- Does it do good?
What is the ethical concept of justice?
- Fairness
-Objectively in evaluating results
What is allocation?
Refers to the prcoess of assigning participants to experimental conditions or groups.
Debriefing
The researcher has to explain simply how and what the study involves the volunteer/s to do, so they understand what is required of them.
Voluntary participation
ALL participants must volunteer to be involves in the study. They cannot be bribed, forced or pressured to be involved.
Withdrawal rights
If at any point in time during the study, a volunteer would like to withdraw, they have right to do so. Once they have left, their data cannot be used or evaluated.
Deception
When debirefing the study to volunteers, the researcher can decieve/mislead the volunteers to think something a little different so they dont interfere with the results and data.
Informed consent
The researcher has to breifely explain their experiment/study to the volunteers in order to recieve their consent to be apart of the study. Including potential risks and the right to withdraw.
Confidentiality
The researcher does not have any right to disclose any information about any volunteer/s for any other purpose than their study. Before disclosing private information or test results, consent must be recieved. Privacy, protection and security.
Controlled experiments
Often used because they allow a researcher to strictly manipulate variables of interest, in a controlled environment and measure their effect on another vairable.
What is a DV
The variables that is used to observe the measure the effects on the IV.
What is an IV
That variable that is manipulated or changed to observe whether it affects another variable and what those effects are.
Aim
The aim of this study is to exlplore——— between ——- and ——-
Graphing conventions
- Title
X and Y labelled with variables - Units of measurment
Mode
Most frequently occuring value in a data set.
Median
The middle value in a data set ordered from lowest to highest.
Mean
Measure of central tendency that describes the numerical average of a data set as a single value.
Measures of variability
- Range
- Standard deviation
Measures of central tendency
- Mean
- Median
- Mode
Percentage change
Formula:
old number-new number / old number multiplied by 100
Outlier
Researchers should be careful for outliers, or values that differ significantly from other values in a data set.
Subjective data
Data that is informed by personal opinion, qualitative descriptions
Objective data
Factual data, not due to personal opinion
secondary data
Data sourced from others prior research, not collected directly by the current researcher.
Primary data
Data collected first hand by a researcher
Double-blind procedure
Procedure in which both participants and the experimenter do not know which conditions or groups participants are allocated to.
Confounding variables
Refers to variables that has directly and systematically affected the DV, apart from the IV. May have been an extraneous variable, that has not been controlled for, or a variable that simply cannot be controlled for. Can only be identified at the end of a study.
What is an extraneous variable?
Variables other than the IV that can cause change in the dv. These variables should be controlled or monitored.
What are non-scientific ideas?
Non-objective, unempirical, imprecise or dogmatic and unverifiable.
What are scientific ideas?
Aim to be objective, utilise and produce empirical evidence and are formed using the methods of science.
Ways to reduce unwanted effects of non-standardised testing and conditions and procedures?
Standardised testing conditions and procedures.
Ways to reduce unwanted effects of situational variables?
Standardised testing conditions and procedures.
Ways to reduce unwanted effects of demand characteristics?
Single blind procedure
Double blind procedure
Standardised testing condtiions and procedures
Ways to reduce unwanted effects of placebo effect?
Single blind procedures-> participant does not know what group they have been allocated to
Placebo
Demand characteristics
Refer to cues in an experiment that may signal to a participant the intention of the study and influence their behaviour
What are participant related variables?
Individual participant differences or subject variables: Characteristics like age, intelligence and socioeconomic status.
Ways to reduce unwanted effects of participant-related variables?
Having a large sample size.
Experimental design choice, making the correct choice for the specific study.
What are order effects?
- Specific limiation of within-subject design
1) Participants oerform better in later conditions: Practise effects
2) Participants perform worse in later conditions due to being tired or bored from computing a prior task:
Fatigue effects
What is a placebo?
A substance or procedure with no active treatment (false drug)
What is the placebo effect?
Placebos are used to avoid the placebo effect. Purely based on expectation.
What are experimenter effects?
Experimenter bias, when the expectations of the researchers affect the results of an experiment.
Ways to reduce unwanted effects of order effects?
Counterbalancing method to reduce order effects that involves systematic manipulation ensures order effects are accounted for, different design.
Ways to reduce experimenter effects?
Single and double blind procedures. Single blind procedure is a procedure in which participants are unaware of the experimental group or condition they have been allocated to.
What are situational variables?
Any environmental factor that may affect the dependent variables such as temp, lighting, weather and time of day.
What are non-standardised instructions and procedures?
The conditions of the test should be the same: Procedure should be same environmental conditions, directions should be the same.
- Non-standardized, introduces unwanted situational variables for either specific participants or entire experimental groups.
D.O.P.P.E.N.S
D= Demand characteristics
O=Order effects
P= Placebo effect
P= Participants related variables
E= experimenter effects
N= Non-standardised instructions and procedures
S= Situational variables
What is stratified sampling?
involves selecting people from the population in a way that ensures that its strata (subgroups) are proportionally represented in the sample.
Stratified sampling advantages:
Most likely to produce a representative sample
Stratified sampling disadvantages:
Time consuming
May not create an entirely representative sample when the sample is small.
Process of stratified sampling:
1) Dividing the research population into different strata based on characteristics relevant to the study.
2) Selecting participants from each stratum in proportion to how they appear in the population.
What is random sampling?
Uses a procedure to ensure every member of the population has the same chance of being selected.
Advantages of random sampling
- Sample generated can be more representitive then convience sampling, reduces experimental bias and fairly representitive if the sample is large.
Disadvantages of random sampling?
May be time consuming and may not create an entirely representative sample when the sample is small.
What is convience sampling
Any sampling technique that involves selection readily available members of the population, rather than using a random or systematic approach.
Advantages of convenience sampling?
Most time effective, can be cost effective.
Disadvantages of convenience samplings?
Most likely to provide a unrepresentative sample for researchers to generate results to the population.
What is a sample?
Subset of the population who participate in a study
What is a population?
Focus of the research, and form which the sample is drawn.
Fieldwork
Refers to any research involving observation and interaction with people and environments in real-world settings, conducted beyond the lab.
Identification
Is a process of recognition of phenomenon as belonging to a particular sets or possibly being part of a new or unique set.
Classification
The arrangement of phenomena, objects or events into manageable sets
Used to create labels
What is a correlational study?
Type of non-experimental study in which researchers observe and measure the relationship between two or more variables without any manipulation of them.
What is a case study?
In-depth investigation of an individual, group or particular phenomena that contains a real or hypothetical situation.
Mixed design
Combines elements of within-subjects and between-subjects designs.
Advantages of mixed design
Allows experimenters to compare results both across experimental conditions and across individuals/participatns/groups over time.
Disadvantages of mixed design
More costly and time consuming. ASks for researchers to be across both methods.
Between subjects design
Individuals are divided into different groups and complete only one experimental condition.
Advantages of between subjects design
May be less time consuming than a within-subjects design. Does not create order effects.
Disadvantages of between subjects design
May need more participants then within-subjects design. Differences between participants across groups can affect results.
Within subjects design
Participants complete every experimental condition.
Advantages of within-subjects design
Results are more likely because of manipulation of the IV than any difference between participants that would occur in small groups. Less people needed as well as good for real world settings.
Disadvantages of within subjects
Order effects
participants dropping out has a greater impact on the study.
