Psych 1111 Flashcards
what is critical thinking
evaluating sources of information and making judgements based on evidence
what are the 4 common sources that people refer to for information?
common sense, superstition and intuition, authority, tenacity
Common sense →
believing that information is correct because it is collectively agreed upon
Superstition and intuition
Gaining knowledge based on subjective feelings → issues with this include lack of knowledge of the source of information (cant evaluate), interpreting random events as causally related, patterns (people like them), caused by priming of attention, observations tend to occur together
Authority
Gain knowledge through authority figures → do they really know what they are talking about?
Tenacity
Gaining knowledge by hearing information so often you accept that it is true
Critical thinker =
scientific thinker
To be a critical thinker you must analyse evidence:
LOSAAR
- Rational
- Analytical
- Logical
- Skeptical
- Open minded
- Able to update your opinion based on the evidence
Assumptions of science
(TVD PEF)
parsimony, empirical, verifications, testability, falsification, determinism
Parsimony
the simplest explanation
Empirical
claims supported by evidence → systematic * The more unusual the claim, the stronger the evidence needs to be.
Verificationism
You must be able to provide evidence that supports your claim.
Testability
must prove claim
Falsification
It must be possible for you to find evidence that refutes your scientific claim. Your scientific claim should allow for the possibility that you are incorrect. Good scientific theories must be able to be falsified
Determinism
is an important assumption of the scientific method. In science determinism refers to the idea that every event in nature has a causes, or causes that account for the occurrence.
Understand the difference between independent and dependent variables
Independent variable is the manipulated variable → is randomly assigned to control for systematic differences → normally has two levels (such as drug and placebo)
Quasi independent variable → variables that the experimenter cannot be randomly allocated → Commonly used as grouping variables
Natural Variables
Country of birth
Biological Sex
Age
CAB
Attribute/person variables
Individual difference variables that fall on a spectrum
Level of risk taking
Anxiety
AIL
Dependent variables
* The dependent variable is the variable used to assess or measure the effects of the independent variable
* Dependent on the independent variable
* Measures a behaviour or response for each treatment condition of the experiment
* The dependent variable is NOT manipulated it is only ever measured
Define and understand the importance of operationalisation
Operationalise = to quantify (measure)
Operationalising variables allows you to specify exactly what you mean in your hypothesis/theory
Operational definition
* Detailed description of the procedures or operations used to measure or manipulate the variables.
* Providing clear instructions about
o Definition of variable
o How it is measured/quantified
* This is important it ensures that the hypothesis is clear.
Identify and apply the steps of the scientific method
Initial/past observations → hypothesis → test → analyse/conclude → update or discard (can go back to the hypothesis step and start again) → theory
Observation
Scientific studies begin with an initial observation.
* A point of interest for further investigation.
* You must be able to find a way to collect observable evidence.
‘Gap in research’
Past observations are important for the scientific method.
* Try to answer questions raised by existing theories.
* Replication is critical. → indicates confidence of results
Hypotheses
* A hypothesis is a very specific statement about the predicted/expected relationship between variables (both variables)
* It is usually phrased in the form: “If ___[I do this]___, then ___[this]___ will happen.”
* A hypothesis usually predicts the effect of a manipulated variable on a measured variable.
* States that a relationship should exist between variables, the expected direction of the relationship between the variables and how this might be measured
Test
The scientific method requires that you can test the hypothesis.
Design an experiment
Use good experimental design
Collect appropriate data
Control as many aspects as possible
Research Methods
Is the experiment reliable?
Are your measures valid?
Analyse and conclude
Consider whether the data supports your hypothesis
Is there sufficient evidence?
Are the results statistically significant?
Are further studies required?
Conclude
Conclusions are the researcher’s interpretation of the evidence
Based on the results of the experiment
Explain the results of the experiment
Update or Discard
The scientific method is dynamic
* Must be able to update your hypothesis when there is a lack of data to support it
* Must be able to discard your hypothesis when the evidence refutes it.
This requires many aspects of critical thinking
* Open to the possibility you are incorrect
* Evaluation of the evidence
* Ability to change your opinion with new evidence
Theories are NOT hypothesis → theory is based on years of work
Theory
* A theory is an organised system of assumptions and principles that attempts to explain certain phenomena and how they are related.
* Many hypothesis are tested and data collected before a theory is formed
* Provide a framework regarding the facts
* Theories can also lead to further questions and hypotheses
Identify and explain the goals of science
The goals of Science
Description → observing phenomena in a systematic manner, needed for prediction
Prediction → make predictions from one variable to another
Explanation → provide a causal explanation regarding a range of variables
Description
* You might want to observe a simple behaviour
* Are people taking “pills” at this festival?
* Might want to investigate something more complex
* Is there a relationship between the type of pill and rates of overdose?
* Need to describe types of pills being consumed
* Need to describe and measure contents
* Need to observe and describe the number of overdoses
Prediction
* Identify the factors that indicate when an event will occur
* Scientific prediction: We are able to use the measurement of one variable to predict the measurement of another variable
* The relationship between variables
* Does X occur with Y?
* Does X change in relationship to Y?
* We are looking at the correlation between two variables
Explanation
* The final goal of science is explanation
* This is the ultimate goal of science
* Is there a causal relationship between X and Y?
* Does X cause Y
* We need to test the causal relationship
* This requires research methods and experimental design
* This requires statistics to evaluate the data
Understand the difference between Pseudoscience and actual science
Science vs. Pseudoscience
* The main difference is that science usually modifies or abandons failed hypotheses/theories when flaws or new evidence have been identified
Unfalsifiable hypotheses/theories
Vague/Unclear/poorly defined concepts
Un-parsimonious hypotheses/theories
Using testimonials
* Need systematic observations
Biased sampling/groups allocation
Placebo Effects/Experimenter bias
* Double-blind control studies
Measurement and Error
- All measurements can be broken into at least two components
- The true value of what is being measured and measurement error
Measured Score = True Score + Error
X = T + e - However, we want: Measured Score = True Score
how do we reduce error?
- Error is reduced with:
Many participants – Individual differences error
Many measurements – Measurement error
Many occasion → able to replicate findings in different contexts - Averages of scores are more reliable than individual scores
Identify and define the types of reliability
Inter-observer reliability
* Degree to which observers agree upon an observation or judgement.
Can be frequency or categorical judgement.
Rating attractiveness.
Scoring rat behaviours.
Coding explanations/descriptions.
* Measure inter-Observer reliability with correlations
Inter-observer reliability
Measure inter-Observer reliability with correlations
* Positive relationship between the scores of each observer
To have high inter-observer reliability we want both observers to agree- Very important for scientific research
* The higher the correlation between observer judgements the more reliable the results are.
Internal/Split-half
Internal Reliability
The degree to which all of the specific items or observations in a multiple item measure behave the same way
* Measuring Intelligence: All the items should equally measure intelligence
High internal reliability shows the entire measure is consistently measuring what it should be
We want more items to measure to reduce error
* Very important that these items all consistently measure the construct we are interested in
Internal Reliability
How can we examine whether multiple items on a test equally measure the same thing?
Divide the test into two halves
Look at the correlation between individuals’ scores on the two halves
Split-Half reliability
- All items of an IQ test should measure intelligence
Need to compare like with like
* Don’t just split in half down the middle (group them and then split down the middle) (we want high correlations)
Look at the correlation between individuals’ scores on the two halves
* High correlation between scores indicates good internal reliability
Internal Reliability
The degree to which all of the specific items or observations in a multiple item measure behave the same way
High internal reliability shows the entire measure is consistently measuring what it should be
We want more items to measure to reduce error
* Very important that these items all consistently measure the construct we are interested in
Test-retest
Test-Retest Reliability
If we were looking at scores on a visual search task, we need the measurement to remain constant over time
Practice effects undermine test-retest reliability
Should counterbalance the order of presentation
Randomly assign people to differ orders
Test-Retest Reliability in practice
Brain Training example
Practice Effects
There is an improvement on scores in the game which indicates poor test-retest reliability
Practice effects – you get better because you do the same task several times
* If I asked you to play Pac-Man every day for 15 minutes and you improved your score, no one would be surprised!
Not a reliable measurement for cognitive improvement
Practice Effects
We have very reliable test-retest measures in these experiments testing the efficacy of brain training
* Shows the tests are reliable measures over time
* Scores on the external measure don’t change after training
* This has been found multiple times
So, these games don’t improve your brain function at all
Replication
The reliability of results across experiments
* Can we replicate the results when all variables and conditions remain the same
* Need clear and detailed method sections
Critical to the scientific method
* Must have evidence from multiple experiments
* More times a result is replicated the more likely it is the findings are accurate and not due to error
Replication Crisis
A lot of published psychological papers couldn’t be replicated
Understand the difference between reliability and replication
It means that if the same study is repeated under the same conditions, it should produce similar outcomes. Replicability, on the other hand, refers to the ability of other researchers to reproduce the results of a study using the same or equivalent methods and data.
Replication
The reliability of results across experiments
* Can we replicate the results when all variables and conditions remain the same
* Need clear and detailed method sections
Critical to the scientific method
* Must have evidence from multiple experiments
* More times a result is replicated the more likely it is the findings are accurate and not due to error
what is validity
- Validity refers to how well a measure or construct actually measures or represents what is claims to.
- Validity relates to accuracy.
- Very important psychology where we often measure abstract constructs
Identify and define the types of validity
Types of Validity
Measurement Validity (measurements are the dependent variable)
* Construct validity
* Content validity
* Criterion Validity → correlations → how valid are the predictions
Concurrent
Predictive
Internal validity → can you make a claim based on these results
* Strength of causal claim
External Validity → can i generalised these claims to the population or environment?
* Population validity
* Ecological (environment) validity
Measurement Validity
Measurement Validity - how well a measure or an operationalised variable corresponds to what it is supposed to measure/represent.
Show that the measurement procedure actually measures what it claims to measure
We use a number of methods to assess the validity of a measurement
o Critical for scientific research
Construct validity
How well do your operationalized variables (independent and/or dependent) represent the abstract variables of interest?
Experimentally: Are you measuring what you think/say you are measuring?
Construct validity = strength of the operational definitions
The strength of your operationalising of variables
Construct validity
For example: Measuring hunger in rats
* Weigh amount of food consumed?
* Speed of running towards food?
* Duration spent at the normal site of food delivery?
* How much they are willing to press a lever for food?
* How would you assess this?
Define hunger
Should relate to manipulations known to produce different levels of hunger e.g. food deprivation
Ideally will be consistent with other measures of hunger
Content validity
Degree to which the items or tasks on a multi-faceted measure accurately sample the target domain
* How well does a measure/task represent all the facets of a construct?
* E.g. IQ tests: 7 questions on a facebook quiz that ask you about mathematics, general knowledge and logical reasoning… can they really adequately represent something like IQ?
Many constructs are multi-faceted and sometimes multiple measures must be used to achieve adequate content validity
Domains = grouping of contents
Need all domains to accurately measure the construct of interest.
Content validity vs internal reliability
Content validity demonstrates that all of the items on a multiple domain measure accurately measure the construct.
* Extroversion scale need all questions to accurately measure extroversion and not another construct
Internal reliability relates to whether the items on a multiple measure domain consistently measure the construct
* Intelligence test: All questions about verbal intelligence should produce consistent score in the same individual
Criterion Validity
Criterion validity measures how well scores on one measure predicts the outcome for another measure.
* The extent a procedure or measure can be used to infer or predict some criterion (i.e. another outcome)
Two types of criterion validity
* Concurrent validity (now)
* Predictive validity (future)
Concurrent Validity
Concurrent validity compares the scores on two current measures to determine whether they are consistent.
* How well do scores on one measure predict the outcome for another existing measure.
* If the two tests produce similar and consistent results you can say that they have concurrent validity
Predict the outcome of a current behaviour from a separate measure
* How many chickens do you follow on instragram? Are you going to buy more chickens now?
Predictive Validity
Scientific theories make predictions about how scores on a measure for a certain construct affect future behaviour
* If the measurement of a construct accurately predicts future behaviour, the measurement has high predictive validity.
Example: Listening to Hip-Hop leads to violent crime
* Compare hours listening to Hip-Hop with number of violent criminal offenses
* Low correlation indicates that listening to hip-hop has poor predictive validity for future violent crimes
Or scores on atar and how successful you will be at university
You want HIGH predictive validity
Internal validity
Internal validity is focused on whether the research design and evidence allows us to demonstrate a clear cause and effect relationship.
High internal validity occurs when the research design can establish a clear and unambiguous explanation for the relationship between two variables
Internal Validity
Relative statement rather than an absolute measure (not a direct way to measure)
* Can we rule out other explanations?
* Are the variables accurately manipulating or measuring the construct?
* Does the research design support the causal claim?
Can’t directly measure internal validity with a correlation.
Crucial to make claims about the causal relationship between variables
External validity
How well a causal relationship hold across different people, settings, treatment variables, measurements and time.
How well we can generalize the causal relationship outside the specifics of the experiment
* Is your sample representative?
* Is the context representative?
* Can results from animal labs generalize to humans?
High external validity occurs when we are able to generalize our experimental findings
Population Validity
Population validity refers to how well your experimental findings can be replicated in a wider population
Aim to have the findings generalize from our experimental sample to the wider population
It is difficult to obtain high external validity in controlled experimental settings
Population Validity
WEIRD population
Western, Educated, Industrialized, Rich, Democratic
Differences in tasks ranging from motivation, reasoning and even visual perception
Can you generalise your results to the wider population?
* Example: Run a test looking at drinking habits – the participants that sign up all happen to be around 22 and female.
Ecological validity
(just means the setting of the experiment → have i thought abou the fact that if it was done in a different environment, would people act the same?)
Ecological validity is how well you can generalise the results outside of a laboratory environment to the real world.
Laboratory experiments vs. real-life settings
* E.g. aggression studies in the lab vs. in real life
Laboratory settings are very controlled and different from real life settings
* People are aware they are under experimental conditions and behave differently
Understand the difference between reliability and validity
Validity = accuracy
Reliability = consistency
Reliability vs. Validity
Reliability: The consistency and repeatability of the results of a measurement.
* My scales at home always consistently tell me that I weigh 55kgs – they are reliable because they produce the same results consistently
Validity: The degree to which a measure or experiment actually measures what it claims to measure
* If my scales are always 5 kg less than my actual weight then it is not a valid measure of my weight (though it would be very reliable if it was always exactly 5 kg off).
Reliability vs. Validity
We want scientific measures to be both reliable and valid
* Reliability demonstrates the measures consistently performs the same way
* Validity demonstrates that the measure actually measures what it claims to measure
* A valid measure that is also reliable – the measure accurately measures what it claims to and does this consistently.
what are the requirements for causality
- J.S. Mills proposed three requirements for causality
Covariation - Is there evidence for a relationship between the variables?
Temporal sequence - One variable occurs before the other
Eliminate confounds - Explain or rule out other possible explanations.
Define and identify confounds
Identify and define confounds
* Third Variable
* Experimenter bias
* Participant effects
* Time effects
Threats to Internal validity
Internal Validity: Strength of Causal Claim
J. S. Mill’s 3 criteria to infer causation
1. Covariation → show relationship between two things
2. Temporal Sequence → one thing causes another
3. Eliminate alternative explanations
Third Variable Problem
Confounds (same as third variable) – Extraneous variable that systematically varies or influences both the independent and the dependent variable
Confound
Confound is a third variable that differs between the groups.
Confounds influence the DV and are not the variable you are manipulating
You may have a different confound in your experimental and control groups
Third Variable Problem
There is a positive correlation between coffee drinking and the likelihood of having a heart attack
Can we conclude that drinking coffee causes heart attacks?
People that are smokers tend to drink more coffee
May be increased job stress
May have poor sleep
Could be why you drink more coffee
Could be due to more coffee
Internal validity threats from the experimenter
Experimenter bias is a confound which undermines the strength of a causal claim
The bias of the experimenter may influence the way a dependent variable is scored
Experimenter may behave in a way that influences the participants and confounds the results of the experiment.
Not always intentional
* Previous knowledge and ideas can create tunnel vision in the experimenter
Example → smart v dumb rats case study
Double blind studies help this
Threats to internal validity: the participant
The way a participant behaves can influence the validity of the results.
Individual differences that are systematic can interfere with the causal relationship you are investigating
Demand Characteristics
* Participants identify the purpose of the study
* Behave in a certain way as a result of identifying the purpose of the study
Unobtrusive observation aids with resolving demand characteristics
Indirect measures also aid this
This also ties into measurement effects
Deception and confederates aid in resolving this
Time related Confounds
Maturation – The effect of time on participants
* Short term: mood, tiredness, hunger, boredom
Deal with this using a number of measures
* Counterbalancing the order of tests
* Control for time of day
* Design experiments of reasonable length
* Include breaks in the experimental design
Maturation: Long term effects
Maturation – The effect of time on participants
* Long term: Age, education, wealth
Difficult to control for long term maturation.
* Only important for longitudinal studies which take place
over many years.
Random assignment and sampling helps to reduce this confound
Define and identify artifacts
Identify and define artifacts
* Mere measurement effect
* History effects
* Selection Bias
Artifacts
Artifacts reduce external validity
Prevents you generalising your results
Unlike a confound, an artifact is something that is ever present in all groups being tested that stays constant
Mere measurement effect
Being aware that someone is observing or measuring your behaviour may change the way you behave.
This is important for external validity as it undermines the ability to generalise lab results to wider population and context.
Similar, to demand characteristics- except that it affects all subjects in the experiment
Not an individual difference variable
History Effects
The effect of a period of time may make an entire sample biased
Example: level of education in Syria
* War zone = limited access to school
* Limited shelter and food
The data is influenced by the moment in time
Can’t generalize these findings to a wider population or different contexts
Selection Bias
Selection bias is where participants volunteer for a study who have a biased interest in the topic of research or the outcome of the study.
You would expect people who really loved beards would be the ones that completed this survey.
Non-response bias
Non-Response Bias is a problem for experiments that involve voluntary sign ups or surveys
People do not respond when they are not interested in something
* You lose a large sample of the population to non-response
bias
This undermines the external validity of the experiment
* Limited population means that the results cannot be generalized to a wider population.
Many Polls and online surveys are subject to this threat
How to Manage selection bias
It is important to use a random sample of the population
Does not eliminate all problems but reduces the likelihood of systematic biases in your data.
Compulsory poll – census
* This reduces sampling bias
* Produces results and data that is more widely applicable
* Less susceptible to biased groups
* Still susceptible to demand characteristics
Understand the difference between confounds and artifacts
Artifacts
Artifacts reduce external validity
Prevents you generalising your results
Unlike a confound, an artifact is something that is ever present in all groups being tested that stays constant
Confound
Confound is a third variable that differs between the groups.
Confounds influence the DV and are not the variable you are manipulating
You may have a different confound in your experimental and control groups
identify and define non experimental methods
- Descriptive & observational research (lowest level of non experimental research) (such as the census) → like creating a database → very large studies to get a lot of data → use this to create further studies. Observational is observing the data → not taking part or manipulating anything (like national geographic → watching animals). Use observational to get an idea about the topic area to then expand on → highest external validity and the lowest internal validity
- Correlational → dont manipulate anything → looking at the relationship between variables/measures → no independent variables → second highest external validity and the scond lowest internal validity
identify and define experimental methods
- Quasi-experiment → use info that we already have to split people into groups → cannot randomly assign participants (such as you cant give people depression to measure depression) (If you want to look at a clinical population). → interested at looking at the different TYPES of participants (sane v the insane) → second highest internal validity and the second lowest external validity (can be generalised more easier)
- True experiment → always has an IV and DV → IV has random assignment → minimises individual differences and third variables → also has a control → highest internal validity but the lowest external validity
Define and identify the characteristics of a true experiment
1) Systematic manipulation of 1 or more variable(s) between or within groups – I.V.
* Guarantee temporal order of cause-effect
* Observe covariation between variables
* Minimize alternative explanations/confounds
2) Random assignment to each condition/group
* Minimise alternative explanations/confounds
- Must have systematic manipulation of the independent
variable - Independent variable = violent video games
- Need to give this an operational definition
- IV is playing violent video games as operationalised by
playing call of duty - Must have a measurement – need to measure the effect
of the IV - Dependent variable – aggression or violence
- Need to provide an operational definition
We have a true experimental design
* Manipulation of the IV gives us confidence in the
cause-effect relationship between violent video
games and aggression
* Control group: Gives us confidence we can
minimise systematic differences
* Random assignment to groups minimises
systematic differences/confounds between our
groups (Internal Validity)
* Random sampling reduces bias in our sample
[Population Validity]
Define and understand random allocation
Random Assignment
* You randomly assign participants to each of the
groups
* Reduces the likelihood of systematic
differences between the participants in the
group which undermine internal validity
* May have some differences but in the long run we
can be sure we don’t have biased assignment
* In the long run over multiple experiments we can
be sure we have eliminated this confound
Identify and understand the difference between random assignment and random sampling.
- Random sampling is an approach to recruiting subjects for
your study - Try to sample different elements of the population proportionally
- More representative sample
- Applies to all forms of research design
- External validity
- Random assignment is an approach to controlling bias in
group allocation - Minimise confounds
- Internal validity
Understand the difference between between and within subject experiments
Within Subject
* Can control third variables in other ways
* Use the same participants in the different conditions
* Repeated measures design/ Within-subjects design
- True experiment – systematic
manipulation of the IV - No random assignment – but
using the same participants for
each task so we remove
individual difference confounds - Sometimes called a repeated
measures design
Advantages
* Can be very powerful – remove the noise
* Powerful in terms of statistics
* Accounts for individual differences
Limitations: Order Effects
* Fatigue
* Practice
* Carry-over
COUNTERBALANCE
Define and identify the characteristics of a Quasi experiment
Characteristics of Quasi-Experiments
Research designs where the researcher has only partial control over the independent variables
Participants are assigned to groups or conditions without random assignment.
Two types of Quasi-Experiments
* Person x treatment
* Natural experiments
Quasi-Experiments are very useful when random assignment is not possible or ethical
Quasi-Experiments
Have dependent variables and sometimes have true independent variables
BUT ALSO HAVE
Quasi-independent variables
Like independent variables except
* Not manipulated by the experimenter
* Random assignment is not possible
Types of Quasi-Independent variables
There are two major types of quasi-independent variable
1. Person/Attribute variable
2. Natural variables
Define and understand the difference between an attribute and natural variable
Person/ Attribute Variables
Individual difference variables
* Can vary along a spectrum
* Can be based on diagnostic criteria
Use these variables most commonly for comparing groups – grouping variables
We can use these to compare any differences on a dependent variable when random assignment is not possible
Person/Attribute Quasi-independent variables
Use of an individual difference variable (essentially a measured not manipulated independent variable)
Note: Must be measured prior to the experiment
Otherwise, there are issues with internal validity
Temporal sequence!
- Need to be sure that we didn’t cause the difference
Attribute variable
Attribute variable: Extroversion vs Introversion
* Have randomly selected participants complete a personality test which measures a range of personality traits.
* Examine the scores and then select a group that score high in extroversion and introversion
The attribute is measured and then participants are split into groups on the basis of their score
How do we split?
Splitting these attribute variables into high and low is a common practice
Not the best method statistically (quite crude)
Median Split
Find mid-point
Advantage:
* Easy
* Get to keep all participants
Disadvantage:
* Participant 10 and 11 are similar…
* Loss of information about unique individual differenc
Natural Variables
Another form of quasi-independent variable is the natural variable
These are variables that are manipulated by nature!
* Sometimes called natural experiments
* Called “acts of god” by insurance companies
For example
* Being in a hurricane
* Living in a warzone
* Country you were born in
* Biological gender
* Age
Natural Variables
Notice that these are all variables that can’t be manipulated by the experimenter or randomly assigned
They have been independently manipulated by Nature.
Natural variable Quasi experiments allow us to look at the effect of war zones, different environments or biological differences
Natural Variables vs. Attribute variables
Can be quite hard to distinguish
* Age – is a natural variable
* Living in a hurricane zone is a natural variable
What about introversion- is this obviously an attribute variable…
* Genes and environment both are usually out of the control of the person! (epigenetics → how genes interact with the environment)
It is not always clear. However, this is not a problem because:
1. They are treated the same way statistically
2. They provide the same kinds of threats to internal validity
Understand and identify a Person x treatment Quasi Experiment
Person/Attribute X Treatment Design
Quasi-Independent variable – measured not manipulated and no random assignment
True Independent variable – manipulated and random assignment
Dependent variable – measured by the experimenter.
Allow us to examine group differences and how they interact with a manipulated or treatment variable
Person/Attribute X Treatment design → my experiment?
Quasi-IV = anxiety levels
* Split participants into severe or moderate anxiety groups based on their scores on a valid anxiety scale
True IV: New treatment v Existing treatment
DV: Reduction of anxiety symptoms
Identify the benefits and limitations of QuasiExperiments
Threats to internal validity
In quasi-experiments the lack of random assignment or controlled/systematic manipulation of the quasiIndependent variable means:
We can never be certain of temporal order of quasi-IV and DV
Third variables -> alternative explanations!
* Pre-existing group differences on other variables
* We can try to match the groups on other characteristics
Why do them?
Higher in External Validity
Match or Patch groups for relevant threats
* Not perfect but can be quite effective for ruling out specific 3rd variables/alternative explanations
Sometimes you can’t manipulate things in the lab
* Not possible to manipulate depression
* Not possible to manipulate whether someone is a psychopath
Patching
Number of different control groups used to try and account for the major threats to internal validity
Forensic control group vs. Psychopath group – similar life situation
Amygdala damage patient compared to subjects with brain damage but in a different region
Community control groups for brain damaged subjects to control for age and IQ differences
Understand correlational designs
How do correlational designs work?
Measuring but not manipulating variables
* Multiple Dependent Variables
* The experimenter is not manipulating anything, just measuring participants
Same issues regarding descriptive studies apply
* Measurement/testing effects
* Question wording
* Random sampling needed to ensure External Validity
Characteristics of Correlational design
WARNING!!!!
* The difference between correlational and quasiexperiments is not always 100% clear
* Correlation ONLY DVS!!! Only measuring NOT manipulating
* Quasi-experiments have more.
Cant use something that is a discrete category for a correlational experiment (has to be continuous variables → like happiness and wealth (on a scale of 1 - 10))
Quasi experiments → split them into groups and distinguish the differences between groups
Correlational studies need continuous variables
Quasi need categorical variables
Define and identify the basic correlations
Positive Correlational Relationship
This relationship is called a positive correlation.
The two variables co-vary in the same direction.
As scores on one variable increase, scores on the other variable increase.
* Also: As scores on one variable decrease, scores on the other variable decrease.
Meaning: As annual salaries increase, the amount of happiness increases.
Or: As annual salaries decrease, the amount of happiness decreases.
Negative Correlational Relationship
This relationship is called a negative correlation.
The two variables co-vary in different directions.
As scores on one variable increase, scores on the other variable decrease.
* Also: As scores on one variable decrease, scores on the other variable increase.
Meaning: As annual salaries increase, the amount of happiness decreases.
* Or: As annual salaries decrease, the amount of happiness increases.
No Correlational Relationship
This relationship is called an No Correlation /uncorrelated.
The two variables do not co-vary.
As scores on one variable increase, scores on the other variable are unrelated.
Meaning: Annual salaries and the amount of happiness are not related
Understand why correlation does not equal causation
Is this a causal statement?
This sounds like a causal statement!
Judging a causal statement: Internal validity
J. S. Mill’s 3 criteria to infer causation
1. Covariation (yes)
1. Temporal Sequence (yes and no)
Hard to establish most of the time
2. Eliminate alternative explanations (no)
Third Variable Problem
Correlation is NOT causation
In informal logic, an argument that tries to suggest that two things are related just because they co-occur or co-vary is a fallacy
* X occurs after Y so they must be related
* X occurs at the same time as Y so they must be related
So, when arguing that correlation implies causation these are the informal logical fallacies that are being committed.
Identify and understand the types of correlational relationships
Correlational relationships
What is the direction of the correlation? You need to be able to identify the temporal sequence of this relationship
* Very hard to do
* Does having more money (x) make you happier (Y)?
* Does being happier (Y) increase your chances of making more money (x)?
* Both of these?
Issue of Reverse Causality
* Cannot determine the direction of the relationship
Indirect Correlational relationships
Indirect correlational relationships occur when there is a variable in between the two variables of interest that is critical to the correlation
Does beauty (x) cause happiness (Y) by increasing wealth (variable z)?
Do cat bites (X) cause depression (Y) by decreasing the amount of time you spend out of the house (Z)?
Third variable Correlational relationships
A third un-measured variable actually causes X and Y and creates the illusion of a correlation between X and Y – confounding variable
Does a failed relationship (variable z) increase the chances of buying a cat and the likelihood of developing depression?
Are wealth and happiness both increased by higher education (variable z)?
Spurious Correlational relationships
Spurious correlations occur when two things appear to co-vary but are not actually related in anyway.
A spurious correlation is different from a third variable correlation as there is no relationship or connection at all between variable X and Y- it just appears this way
* “correlation does not mean causation”
Illusion → can be called an illusionary correlation
Identify and understand the confounds and limitations of correlational research
Sources of confounds in correlational designs
Person Confounds – Individual differences that tend to co-vary
* For example: Depression and feelings of loneliness (and thus the desire for a cat)
* Depression and anxiety
Environmental Confounds - Situations that cause multiple differences
* For example: coming to UNSW can increase knowledge and anxiety
* Listening to your lecturer can simultaneously increase boredom & frustration.
Methodological Sources of Confounds
Operational Confounds – A measure that measures multiple things
For example: Correlation between impulsivity and poor decision making
Definition of impulsivity: a tendency to act on a whim, displaying behaviour characterized by little or no forethought, reflection, or consideration of the consequences.
So, poor decisions are part of the definition, therefore they are correlated by definition!
Limits of correlational research
Correlational studies look at the relationship between measured variables
* Can establish co-variation
* Cannot establish temporal sequence effectively
* Cannot eliminate alternate explanations effectively
* Low in Internal Validity
Confounds can arise due to:
* Individual differences
* Environments
* Operational definitions
Define the difference between nonexperimental and experimental research
Non experimental:
Descriptive
correlational
Low internal validity but high external validity
No manipulation
Measurement only
Experimental:
Quasi → non random assignment
True experiment → Random assignment
High internal validity
Low external validity
Descriptive Research
No Independent Variables
Only Dependent Variables
Aim is to measure and describe
* Not to explain
* One of the 3 aims of science
* [description, prediction, explanation]
Can be thought of as only looking at a single dependent variable
Descriptive Research
Aims to simply describe what is occurring in a certain context.
Alfred Kinsey
Interested in sexuality
What percentage of the population is homosexual?
Based on a large survey Kinsey questioned the label of homosexuality
Found that this label was inadequate
Survey Methods
majority of descriptive studies are conducted by surveys
Survey benefits
Limited data from large samples
* Opposite to case studies
Address questions of “how many”, “how much”, “who” and “why”
Advantages:
* Quick and efficient
* Very large samples
* Obtain public opinion almost immediately
* Simple to use
Observational Research Overview
Usually good for external validity
Terrible for internal validity (by themselves)
Observational studies allow for observation in the real world
Participant observation can lead to issues of experimenter bias
Longitudinal and Cross-Sectional designs
* Cannot manipulate variables but can get a sense of behaviour over time or across groups
Define and identify the types of descriptive and observational research
Descriptive and Observational Studies (can use the terms interchangeably)
(none of these have an independent variable)
Types of descriptive and observational research
Case Study
* Single subject
Descriptive research
* Describe and measure
* No independent variables
Observational research
* Observe subjects
* No independent variables
Understand the benefits and limitations of descriptive and observational research
Dangers of Non-Random Sampling
You gain a representative sample by taking a random sample of the population
Surveys often have response bias
This is critical as it reduces the generalisability and the results of the survey.
Naturalistic Observation
Advantages
* necessary in studying issues that are not amenable to
experimentation
* extremely useful in the initial phases of investigation
Disadvantages
* Cannot determine cause-effect relations
* No internal validity
* Very time consuming
* Observed aware of observer?
Hawthorne Effect
Participant Observation
Advantages:
* It can be used in situations that otherwise might be closed to scientific investigation
Disadvantages:
* The dual role of the researcher maximizes the chances for the observer to lose objectivity and allow personal biases to enter into the description
* Time consuming and expensive
Longitudinal Research
Longitudinal research – follow the same participants across a long time period
Advantages
* Genuine changes and stability of some characteristics observed
* Major points of change observed
Disadvantages
* Time consuming and expensive
* Participant attrition – threat to validity
Cross- Sectional Research
Take groups from different points in time to get a crosssection of the community
Advantages
* Relatively inexpensive and less time consuming
* Low attrition rate
Disadvantages
* Cannot observe changes in individuals
* Insensitive to abrupt changes
* Age-Cohort effects
Relevance of Statistics in Psychological Research
Data-Driven Insights
Statistics are essential for extracting meaningful insights from the vast amounts of data collected in psychological studies.
Informed Decision Making
Statistical analysis helps psychologists make evidence-based decisions and draw reliable conclusions about human behavior and cognition.
Collaboration and
Replication Rigorous statistical methods enable psychological research to be shared, replicated, and built upon by the scientific community
The Ethical Imperative: Why Understanding Statistics Matters
Ethical Data Practices
Understanding statistics is crucial for
ethically collecting, analyzing, and
presenting data. It helps avoid
misrepresentation and unintended
biases.
Informed Decision-Making
Proficiency in statistics empowers
psychologists to make well-founded,
evidence-based decisions that
positively impact research and clinical
practice.
Transparency and Accountability
Robust statistical knowledge fosters
transparency, allowing psychologists
to communicate findings clearly and
be accountable to research
participants and the public.
Advancing the Field
Mastering statistics is essential for
pushing the boundaries of
psychological research and
contributing to the ethical progress of
the discipline
Implications of Misreporting
Overgeneralization
Misleading one-size-fits-all impression of therapy
effectiveness.
Patient Harm
Wasted time on ineffective treatments.
Research Mistrust
Damages credibility of psychological studies.
Ethical Responsibility
Researchers must present complete picture, including
Limitations.
Measures of Central Tendency
Mean
The arithmetic average of a
set of values. Calculated by
summing all the values and
dividing by the total
number of values.
Median
The middle value when the
data is arranged in
numerical order. Useful for
skewed distributions where
the mean may not be
representative.
Mode
The value that occurs most
frequently in the dataset.
Can identify the most
common or typical value.
When to Use Each
The mean is most
commonly used, but the
median or mode may be
more appropriate
depending on the
distribution and research
Goals.
Measures of Dispersion: Understanding Variability
Range
The difference between the highest
and lowest values in a dataset,
indicating the overall spread.
Variance
The average squared deviation from
the mean, capturing the dataset’s
overall dispersion.
Standard Deviation
The square root of variance,
providing a more intuitive measure of
the average deviation.
Interquartile Range
The difference between the 75th and
25th percentiles, describing the
middle 50% of the data.
what are histograms and bar charts?
Histograms and bar charts are powerful tools for visualizing
the distribution of data. Histograms display the frequency of
values, while bar charts compare the magnitudes of
different categories.
These visualizations help identify patterns, outliers, and the
overall shape of the data - crucial for gaining insights and
communicating findings effectively.
whats a box plot
Boxplots offer a concise yet powerful way to visualize the distribution of
data. They display the median, interquartile range, and any outliers,
providing valuable insights into the spread and symmetry of a dataset.
Analyzing the boxplot can reveal key characteristics such as the presence
of skewness, the extent of variability, and the identification of unusual
data points. This visual tool is especially helpful for quickly comparing
data distributions across different groups or conditions.
whats a scatterplot?
Scatterplots allow us to visualize the relationship between two variables.
The pattern of data points reveals the strength and direction of the
correlation - whether the variables are positively, negatively, or not
correlated.
Analyzing scatterplots provides insights into the nature of the
relationship, highlighting potential trends, clusters, and outliers. This lays
the groundwork for deeper statistical analysis to quantify the correlation
coefficient and determine its significance.
Importance of Data Cleaning and Preparation
Identifying Errors
Thoroughly inspect your data for
missing values, outliers, and
inconsistencies that could skew
your analysis.
Handling Missing Data
Decide on appropriate methods to
address missing information, such
as imputation or exclusion, to
maintain data integrity.
Standardizing Formats
Ensure all data is in the correct
format and units to enable
accurate comparisons and
calculations.
Transforming Variables
Apply necessary data
transformations, such as
logarithmic or square root, to meet
statistical assumptions.
Handling Missing Data: Strategies and Considerations
Imputation
Replace missing values with
estimates based on patterns in the
existing data, such as mean or
median substitution.
Listwise Deletion
Remove any cases with missing
data, but this can reduce statistical
power and introduce bias if the
missingness is not random.
Multiple Imputation
Generate multiple plausible values
for each missing data point to
account for uncertainty, then pool
the results.
Analysis of Missingness
Investigate the patterns and
mechanisms behind missing data
to select the most appropriate
handling method.
Interpreting Descriptive Statistics: Beyond the Numbers
Visualizing the Data
Graphs and charts can bring descriptive statistics to life, revealing
patterns, outliers, and relationships that may not be evident in raw
numbers alone.
Contextual Interpretation
Understanding the real-world implications of descriptive statistics
requires considering the study design, sample characteristics, and
potential biases.
Practical Significance
Statistical significance alone does not necessarily equate to practical
or clinical significance. Evaluating the magnitude of effects is key.
Ethical Considerations in Data Presentation
Transparency
Ethical data presentation
means being transparent
about the source,
methods, and limitations of
the data. Hiding key details
can mislead or manipulate
the audience.
Avoiding Bias
Carefully consider how
data is visualized and
framed to ensure it does
not introduce unconscious
biases. Selective
highlighting or omission
can skew interpretation.
Context Matters
Ethical practice requires
providing appropriate
context to help the
audience understand the
full picture. Isolating data
points without broader
context can be misleading.
Responsible Reporting
Researchers have a duty to
report findings accurately
and avoid sensationalizing
or exaggerating results.
Honest, objective
presentation builds trust in
the scientific process.
Avoiding Common Pitfalls in Descriptive Statistics
Misinterpreting Visualizations
Ensure proper understanding of graph
types and their limitations to avoid
drawing incorrect conclusions from
descriptive data.
Choosing Inappropriate Analyses
Matching the right descriptive statistic
to the research question is crucial to
obtain meaningful and ethical insights.
Data Entry Errors
Meticulous data cleaning and
verification are essential to ensure the
accuracy of descriptive statistics and
Visualizations.
Practical Applications of Descriptive Statistics in Psychology
Research Design
Descriptive statistics are essential for
planning studies, determining sample sizes,
and interpreting research findings.
Psychological Assessment
Measures of central tendency and
variability help clinicians understand client
test scores and make informed decisions.
Intervention Evaluation
Descriptive stats allow psychologists to
track progress, identify areas for
improvement, and demonstrate program
effectiveness.
Data Visualization
Charts and graphs based on descriptive
statistics enhance communication and
improve understanding of psychological
Phenomena.
- How does a scientist approach thinking differently from everyday thinking?
Answer: Scientists rely on objective analysis, systematic observation, and evidence-based conclusions, avoiding assumptions or subjective beliefs. They apply skepticism, empiricism, and critical thinking to form judgments.
- Why is critical thinking important in evaluating scientific evidence?
Answer: Critical thinking allows individuals to objectively analyze and evaluate evidence, assess credibility, recognize biases, and form well-supported conclusions. Without it, people might accept information based on authority, intuition, or tenacity without validating it.
- What are some sources people commonly rely on for information? Why might these be unreliable?
Answer: Common sources include common sense, superstition, intuition, authority, and tenacity. These sources are unreliable because they often lack systematic evidence, are based on personal bias, or rely on repetition rather than factual support.
- Explain the issues with relying on ‘folk wisdom’ or common sense.
Answer: Folk wisdom often includes contradictory statements (e.g., “Absence makes the heart grow fonder” vs. “Out of sight, out of mind”) and lacks systematic evidence, leading to unreliable or biased conclusions.
- What is parsimony, and why is it important in scientific research?
Answer: Parsimony, or Occam’s Razor, suggests choosing the simplest explanation with the fewest assumptions when multiple hypotheses predict the same outcome. It prevents unnecessary complexity and focuses on the most likely solution.
- What does ‘extraordinary claims require extraordinary evidence’ mean?
Answer: This principle, associated with Carl Sagan, means that highly unusual or improbable claims need very strong and compelling evidence. For instance, seeing a celebrity in public might only need a photo, but alien encounters require extensive, credible proof.
- Describe the concept of verification in scientific research.
Answer: Verification involves providing observable, confirmable evidence to support a claim. For a hypothesis to be scientifically valid, there must be evidence that can be consistently observed by others.
- What is falsification, and why is it important in science?
Answer: Falsification, proposed by Karl Popper, is the idea that scientific claims must be able to be proven wrong. A hypothesis should allow for the possibility that it might be incorrect, encouraging rigorous testing and honest evaluation.
- Give an example of how people naturally seek confirmatory evidence rather than falsification.
Answer: People tend to search for information that supports their views, such as googling “Does homeopathy work?” instead of “Evidence that homeopathy doesn’t work.” This confirmation bias prevents objective analysis.
- Why might relying on authority figures for information be problematic?
Answer: Authority figures can have biases, and they may not be experts in the specific area of inquiry. Evaluating evidence even from authorities is necessary to avoid misinformation or unsupported claims.
Q1: What are the four key ethical principles in psychological research?
A1: The four key principles are:
Do no harm
Informed consent
Protection of privacy
Valid research design
Q2: Explain the “Do No Harm” principle in research ethics.
A2: This principle ensures that researchers avoid causing physical, mental, or emotional harm to participants. It aligns with the principle of non-maleficence and emphasizes the importance of minimizing harm and discomfort in research.
Q3: Why is informed consent essential in psychological research?
A3: Informed consent ensures participants are aware of the study’s nature, potential risks, and their right to withdraw without consequence. It is a legal and ethical requirement to respect participants’ autonomy.
Q4: How does a “valid research design” contribute to ethical research?
A4: A valid research design ensures that the study has the potential to provide meaningful results, justifying any risks involved. Ethical panels evaluate this design to weigh the cost-benefit ratio and ensure ethical standards are met.
Q5: Describe the unethical practices observed in WWII German medical trials on concentration camp prisoners.
A5: The Nazi medical trials included experiments to create immunity to tuberculosis, where Dr. Heissmeyer injected live tuberculosis bacteria into subjects’ lungs and removed lymph glands. Dr. Joseph Mengele also conducted inhumane twin studies, including injecting chemicals into eyes and attempting to create conjoined twins.
Q6: What ethical dilemma arises from using data obtained from unethical experiments like those of WWII?
A6: Although the methods were unethical, some argue that the data might hold value for modern medicine. This raises a dilemma about whether using this data is justified if it has potential life-saving applications.
Q7: How did the Nuremberg Trials contribute to modern ethical guidelines in research?
A7: The Nuremberg Trials exposed the Nazi war crimes, including unethical human experimentation. This led to the establishment of the Nuremberg Code, a set of ethical principles for human research that strongly influenced later guidelines.
Q8: Name three major ethical bodies for psychological research and where they are located.
A8:
American Psychological Association (APA) – USA
British Psychological Society (BPS) – UK
Australian Psychological Society (APS) – Australia
Q9: Summarize the ethical dilemma in Henle & Hubbell’s (1938) study on egocentricity in conversation.
A9: The study involved unobtrusive observations, raising issues around informed consent as participants were unaware they were being observed. Although no physical harm was done, the lack of consent and potential discomfort make it ethically questionable.
Q10: What were the ethical issues in Zimbardo’s (1973) Stanford Prison Experiment?
A10: Ethical concerns included psychological harm, as participants experienced significant stress and distress. There was limited informed consent since participants didn’t expect to be arrested at home, and privacy was compromised as arrests happened publicly.
Q11: Why was deception used in Milgram’s (1963) obedience study, and what ethical concerns did it raise?
A11: Deception was necessary to test genuine obedience, but it compromised informed consent as participants didn’t know the study’s true nature. The study caused psychological distress, raising concerns about harm and whether the deception was justified.
Q12: What are the three “Rs” in animal research ethics?
A12: The three “Rs” are:
Replacement: Use alternative methods if possible.
Reduction: Minimize the number of animals used.
Refinement: Improve procedures to reduce suffering.
Q13: Discuss the main ethical dilemma associated with animal research.
A13: The ethical dilemma centers on whether the potential benefits to human health justify the harm to animals. Although animal physiology often mirrors human systems, critics argue that ethical standards should protect animal welfare, while supporters focus on the value of research outcomes.
Q14: Define scientific misconduct and name four forms it can take.
A14: Scientific misconduct refers to unethical practices in research. The four main forms are:
Plagiarism: Using others’ work without credit.
Conflict of Interest: When personal gain influences research outcomes.
Fabricating Data: Making up data that didn’t exist.
Falsification of Data: Manipulating or selectively reporting data.
Q15: Give an example of a famous case of fabrication in psychological research.
A15: Diederik Stapel, a Dutch psychologist, fabricated data in at least 30 published studies. His actions significantly impacted the credibility of social psychology research.
Q16: Explain how conflicts of interest can bias research findings with an example.
A16: Conflicts of interest occur when a researcher’s personal or financial gain could skew results. For instance, Coca-Cola funded studies suggesting sugar doesn’t contribute to obesity, raising questions about the impartiality of these findings.
Q1: Why is data analysis considered an ethical issue in psychology?
A1: Data analysis and reporting require transparency to avoid misleading interpretations. Ethical data analysis ensures accurate representation of results, respects participant confidentiality, and avoids manipulation or selective reporting that could misrepresent findings.
Q2: Explain the difference between statistical significance and clinical significance.
A2: Statistical significance shows patterns in the data, indicating whether observed effects are likely due to chance. Clinical significance, however, considers if a treatment has a meaningful impact on participants, addressing practical implications beyond mere patterns.
Q3: Why is transparency important in data analysis, particularly regarding the replicability crisis?
A3: Transparency helps other researchers replicate studies and achieve similar results, which is crucial for scientific credibility. The replicability crisis—where studies often fail to replicate—highlights the need for clear data reporting and honest disclosure of limitations.
Q4: Outline the main steps in the data analysis process.
A4: The key steps are:
Collect: Gather data from surveys, experiments, or observations.
Organize: Use tools like Excel to structure data.
Analyze: Apply statistical methods to identify patterns.
Interpret: Draw conclusions to answer the research question.
