Psychology- Test 2 Flashcards
What are the ethics in psychology research I need to know?
- Informed consent
- Voluntary participation
- Withdrawal rights
- Confidentiality
- Deception in research
What is a focus group?
A group in which a researcher asks members about their ideas, perceptions, opinions etc. in a setting more natural than an interview.
What is a convenience sample?
A sample of the population that is readily available. It is unlikely to be representative.
What is predictive validity?
The extent to which a measure can predict other attributes or behaviours thought to be related to the construct.
What is the “construct” in psychology?
A psychological construct is a label for a domain of behaviours.
What is reliability?
Reliability refers to the consistency of a measure. A test is considered reliable if we get the same result repeatedly.
What is reliability?
Reliability refers to the consistency of a measure. A test is considered reliable if we get the same result repeatedly.
What is qualitative data?
Information that is not represented in numbers.
What is a control group?
A group that is as similar as possible to the experimental group on all variables other than the independent variable.
What is an operational hypothesis?
A prediction of the relationship between the independent and dependent variable, in which the variables have been operationally defined (i.e. stated in terms of how they will be measured).
What
How do you operationalise a variable?
To define the variable so that it can be measured quantitatively or qualitatively.
E.g. Memory by the number of items correctly recalled from a list of 10 items after one minute.
E.g. Aggression by the number of times a student physically hits another person with the intention to cause harm.
What is a dependent variable?
The variable that is measured to determine its relationship to the changed independent variable.
What is an independent variable?
The variable that manipulated or changed by the researcher, not by other variables. This affects the dependent variable.
What is the scientific method?
A systematic approach to planning and conducting research to provide empirical evidence for conclusions reached.
What is a non-experimental method?
A scientific approach to research whereby the researchers do not manipulate variables- the factors of interest. They observe and describe variables in the world around them and note their relationships to one another. E.g. researcher observes child to see if they select partners of same ethnic group.
Cannot determine causation
What is a non-experimental method?
A scientific approach to research whereby the researchers do not manipulate variables- the factors of interest. They observe and describe variables in the world around them and note their relationships to one another. E.g. researcher observes child to see if they select partners of same ethnic group.
Cannot determine causation
What are experimental methods?
A scientific approach to research whereby the researcher tests whether changes in one variable or group of variables have an effect on another variable.
Determines causation with a specified level of certainty.
What is a population in research?
The group to whom we want to apply our research findings.
What does internal consistency mean?
Within a test, different questions or activities are measuring the same construct.
Split half method is evidence of internal consistency.
What is the split-half method?
Evidence for internal reliability.
E.g. in 20 item scale to measure depression, people’s responses in first 10 items should be consistent with their responses with the second 10 items. They should measure the same construct.
E.g. if first 10 indicated that they were depressed, and second 10 indicated that they were not, no internal reliability.
What is test-retest reliability?
Invokes comparing how people performed on a test at one time with how they performed on it sometime later.
What is validity?
The extent to which the experiment actually test what it is meant to test.
What are the types of validity?
- Face validity
- Construct validity
- Concurrent validity
- Predictive validity
What is face validity?
The extent to which the test appears to be measuring what it claims.
E.g. questions on IQ test relate to how we organise, remember and use information.
What is construct validity?
The extent to which the test items are in keeping with the constructs on which the test was based.
E.g. personality test would be expected to relate to three constructs believed to make up personality- extraversion, neuroticism, and psychoticism.
Should not find items relating to constructs not part of theory.
What is concurrent validity?
Comparing people’s performances on scale interested in with performance on one we already know a lot about.
Look at strength of relationship between them.
E.g. new IQ test developed- compare it to an old one.
What is predictive validity?
Extent to which it can predict other attributes or behaviours thought to be related to constructs tested.
E.g. employers issuing personality test to determine conscientiousness want to know that employee will be timely.
What are non-experimental methods?
Researchers do not manipulate variables.
They observe and describe variables in world around them and not their relationship to one another.
Not causational.
- Case studies
- Correlational study
- Archival research- information previously collected by others, but is in a form that allows systematic studies.
- Surveys
What is a correlational study?
Not causation. Use statistical technique of correlation to determine whether there is relationship between variables and how strong it is.
What is archival research?
Information previously collected by others, but is in a form that allows systematic studies.
E.g. official records of hospitals.
What are the steps of the scientific method?
1) Identifying a research question
2) Formulate a hypothesis
3) Design the experiment
4) Collect the data (conduct)
5) Analyse the data
6) Interpret the data
Record findings
What makes continuous data different?
Values can occur in between whole numbers.
E.g. Time, weight, height
If it has units, it is continuous.
What is discrete data?
Data that is counted. E.g. Cannot have 2.5 cars counted.
What is different about psychology graph titles?
Title must be operationalised
• Must have population
• How dependent variable is measured (e.g. Questionnaire, heart rate)
What is a histogram?
Looks like a column graph, but bars are touching each other- indicates continuous data.
Each bar of continuous data is in a RANGE.
What is a frequency column graph?
Shows frequency.
Points are plotted in the middle of each range.
Must draw straight line down to x-axis on either side.
What is a pie chart?
Plotting proportion of something.
Proportion represented by a “slice”.
What is an example of an operationalised title?
What do diagrams show?
Relationships between the variables.
Does not show causation.
What do the directions of arrows represent in a diagram?
The factors that are hypothesized to have an effect on behaviour.
How do you analyse quantitative data?
Mean, median, mode, range.
What are descriptive statistics?
Descriptive statistics are used to summarise, organise and describe data obtained from research
Measures of central tendency: mean, median, mode
What are the strengths and limitations of the mean?
Strength: most sensitive measure of central tendency as it takes account of all scores.
Limitation: Heavily affected by outlier- outlier can skew mean and give incorrect interpretation of relationship between variables. Therefore, could be unrepresentative.
What if the numbers required to find the median are even?
Take the average of 2 numbers.
What are the strengths and weaknesses of the median?
• Strength:
○ Unaffected by outliers in data set
• Limitation:
○ Does not take account of most scores so it can be misleading
What are the strengths and limitations of mode?
• Number occurrent most frequently
• Count which occurs most
• Strengths:
○ Unaffected by extreme scores
○ Useful when data are in categories
§ Frequencies
§ Why?
• Limitations:
○ Can have more than one mode
Unrepresentative measure
What is a measure of dispersion (spread)?
Tells us how close data is to mean
E.g. range
Good indicator of variability
What is informed consent?
Whenever possible, must be obtained.
Requires written consent.
Researcher outlines what research is about, then asks for consent.
Presumptive consent- group of people ask to infer whether it is ok
• Researcher informs BEFORE experiment: ○ Procedures ○ Purpose ○ Expectations ○ Their rights § Withdrawal rights ○ Foreseeable risks ○ Written consent
What is confidentiality?
Data gained must be kept anonymous unless participant gives full consent. No names must be used in lab report
• Personal details and results private ○ Kept secure e.g. Password protected file ○ Cannot be identifiable when results public ○ Cannot be disclosed to anyone without consent § Under 18- guardian
What is voluntary participation?
• Willing
• No pressure or coercion
• E.g. Good grade
• No threat of negative consequences
What are withdrawal rights?
Participants should be able to withdraw themselves and their data from the investigation.
• No negative consequences
• Researcher can’t pressure them
○ Even if detrimental
What is deception in research?
Where participants are misled or wrongly informed about the aims of the research.
Types: deliberate (e.g. deceptive instructions) or deception by omissions (e.g. creating ambiguity)
Sometimes necessary to avoid demand characteristics (clues in experiment which leads participant to know think they know what experimenter is looking for).
Should avoid unless there is no alternative- judged by expert
Deception must not cause stress- must consider cultural factors
Benefit must outweigh harm
What is stratified sampling?
• Can be used in place of random sampling when target population is large
• Researcher identifies the different types of people that make up the target population and works out the proportions needed for the sample to be representative
○ Population is broken down into “strata” or groups, based on characteristics they share
• Random method pulls out same proportion, but smaller subset
What are the disadvantages of stratified sampling?
• Extremely time consuming, therefore rarely used as researcher must collect information about population’s characteristics before sampling
• Requires access to strata and participants within each stratum
Must be carried out within each strata to be representative
What are the advantages of strata sampling?
• Highly representative of target population, and therefore results applied to target populations
• Ensures subgroups (strata) represented in same proportions as in populations
Randomly selected
What is convenience sampling?
• target population available at time and willing to participants
• Participants obtained by asking members of population of interest who are easily and conveniently available, if they would take part in research
• For example:
○ E.g. Would be selecting a sample of students from those coming out of the library
What are the disadvantages of convenience sampling?
Participants are likely to be highly biased (not able to be generalised to population)
What are the advantages of convenience sampling?
• Does not require forward planning i.e. Quick
• Quick to administer
Useful in pilot study to check whether hypotheses are on track and materials & techniques are appropriate
What is random sampling?
• Planned & systematic approach
• Everyone in the entire target population has an equal chance of being used in the study
• E.g.
○ Pulling a name out of a hat/container
○ Allocating a number to each person in the population and then using a random number generator to select the sample
What are the advantages of random sampling?
• Unbiased- sample should represent target population and eliminate sampling bias
Results can be applied to the population
What are the disadvantages of random sampling?
• Very difficult to obtain a complete list of all possible participants or to collect data from mall selected
Expensive i.e. Time, effort, money
What are extraneous variables?
any variable that you’re not investigating that can potentially affect the outcomes of your research study.
difficult to determine a relationship between IV and DV
What is a confounding variable?
extraneous variable that was not controlled during experiment and is believed to have effected dependent variable
What are the types of extraneous variables?
• Participant differences
○ Differences amongst research participants e.g. Moods, personality, ability, memory
• Experimenter effects
○ Differences in how experimenter treats participants in experimental & control groups
• Situational variables
○ Differences in how participants react to experimental environment itself
What is a control group?
Creates point of comparison to experimental group- ensures IV actually causes effect.
Control groups: exposed to absence of manipulated variable
○ E.g. Does exercise reduce depression symptoms?
○ No exercise
must be treated as similarly as possible
What is an experimental group?
• Experimental groups: exposed to changes or manipulations in experiment (IV), to see what happens to other variable (DV)
○ E.g. Does exercise improve depression symptoms?
○ Experimental group exposed to exercise
How are experimental and control groups chosen?
○ Random
○ Matched participants
○ Used when extraneous variable could confound
○ Choose subject with same behaviour/physical attitude
○ E.g. Age and memory
• Repeated measures:
○ Used when researchers believe individual participant differences must be controlled
○ Participant is exposed to experimental & control conditions
How is quantitative data collected?
Experiments.
Controlled observations.
Surveys: paper, kiosk, mobile, questionnaires.
Longitudinal studies.
Polls.
Telephone interviews.
Face-to-face interviews.
Discrete or continuous
What are sources of error in data?
Human error – mistakes e.g. reading a scale incorrectly.
Random error – unpredictable that occur in all experiments e.g. no measurement can be made with absolute precision.
Systematic error – occur as a result of poor design of the method.
Participant sources – Hawthorne & placebo effect
Experimenter sources – experimenter effect
What is the Hawthorne effect?
Participant error
Tendency in some individuals to alter their behaviour in response to their awareness of being observed
Leads to incorrect conclusions about effect of changes in IV on DV
What is the placebo effect?
Placebo Effect
Can occur in experiments designed to evaluate the effectiveness of a drug if the procedure is not designed correctly.
Placebo Effect occurs when there is a change in a participant’s behaviour due to their expectation about the treatment.
How is the placebo effect minimised?
A placebo- neutral substance presented same way, but no active ingredients.
What is a single-blind test?
Single-Blind Procedure- half the participants are given the experimental treatment and the other half are given a placebo
Neither participant know which group they are in.
Any differences in the effects of the drug should be due to the drug itself rather than to the beliefs about the treatment.
What is the experimenter effect?
Occur when an experimenter influences participant’s behaviour.
Experimenter Effect is a change in behaviour by the participant that is due to the behaviour of the researcher.
How is the experimenter effect minimised?
Double-Blind Procedure.
Neither the participant or the person collecting the data knows which group a participant is in.
An independent person is required to allocate participants to the experiment or control group.
After the data is collected, the researcher is made aware of who was in which group.
What are the measures of central tendency?
Mode, median, mean
What are measures of dispersion?
Range
Why do random errors occur?
No human measurement can be made with 100% accuracy.
When does experimenter become aware of who is in which group?
After the data is collected.
What type of errors are participant and experimenter errors?
Systematic error- error in method. Have not considered blind tests, etc.
What are the values of research?
- Justice
- Beneficence
- Research merit and integrity
- Respect
Case studies strengths and limitations:
Strengths: Gain very specific information about an occurrence or phenomenon
Limitation: time consuming
Observation
An individual watches & records another individual or a group of people based on behaviour they see.
Strength: Eliminate extraneous variable of artificiality (effect of an unnatural environment)
Limitation: Rely on the observer’s interpretation of events (observer bias)
Self-reports
• Individuals are asked to comment on their own thoughts, emotions and beliefs by answering a series of questions on a topic
• Include: surveys & questionnaires, interviews (open-ended/fixed response, structured/unstructured), focus groups
Self-reports strengths and limitations
Strengths: Gain detailed information (except surveys & questionnaires)
Limitations: • Difficult to compare data between participants due to subjectivity of the data
Social desirability bias
Scattergram
Scattergram:
- Used to represent correlational data
○ Never join the dots
Correlational studies
- Correlation means association- measure of extent to which two variables are related
- Use correlation to determine whether there is relationship between variables AND how strong relationship is
- 3 results of correlational study:
○ Positive correlation, negative correlation, no correlation - Correlation does not mean causation
Why are non-experimental methods used, and what are they?
- Times during which a variable cannot be practically or ethically manipulated
- Occurs when researchers do not manipulate variables
○ They observe (watch & record) and describe variables in the world around them and note their relationship to one another
§ Can lead to observer bias
○ Relies on:
§ Observations and interactions to reach conclusions
○ Non-experimental cannot determine cause and effect
§ Can give correlation, not causation
- Occurs when researchers do not manipulate variables
Case studies
an in-depth, direct behavioural observation of a single person or small group of people e.g. Phineas Gage
Usually conducted via longitudinal study
○ Advantages: § Provides detailed (rich qualitative) information ○ Limitations: § Researcher's own subjective feeling may influence case study (researcher/observer bias)
Time-consuming and expensive
Surveys
- Gather self-reported data from participants who (ideally) have been randomly selected
- Usual form: questionnaire, in which predetermined answers are provided as alternatives
- Advantages
○ Cheap
○ Quick to administer - Limitations:
○ Lack detail
Responses fixed-> less scope for respondents to answer authentically
Qualitative methods of data collection
- Qualitative data: data that describes, in detail, any changes in behaviour
○ Can’t graph it- Prone to observer bias: person chooses what to write down
Archival research advantages and disadvantages
○ They can combine data from different sources
§ Advantage:
□ More cost and time effective
§ Limitations:
□ No full disclosure as to how data was retrieved
® Lack of ethics
Conclusion
a statement that addresses the original aim or research question
- Was original hypothesis supported or disproven?
- Include reference to the original hypothesis
- State whether results were statistically significant- do we know that results are due to the independent variable?
*Is it statistically significant to derive relationship between IV & DV - State whether results of study could be generalised to population
Generalisation criteria
A generalisation should only be made if all of the following criteria are met:
- Results are statistically significant (using inferential statistics)
- Sample is representative of the population
- Method of sampling is appropriate
- (Where possible) extraneous & confounding variables have been controlled
Generalisation criteria
A generalisation should only be made if all of the following criteria are met:
- Results are statistically significant (using inferential statistics)
- Sample is representative of the population
- Method of sampling is appropriate
- (Where possible) extraneous & confounding variables have been controlled
Generalisation criteria
A generalisation should only be made if all of the following criteria are met:
- Results are statistically significant (using inferential statistics)
- Sample is representative of the population
- Method of sampling is appropriate
- (Where possible) extraneous & confounding variables have been controlled
