Doing Paychology Flashcards
What is psychology
The “scientific” study of the “mind”
What is the “science” of psychology
- the observation of facts:
- they are limited as we can’t see all the facts, forced to make a judgment
- observations interpreted, (problem of objectifying as facts can be disputed
- verifiability (by observation) inadequate
What is the issue of “science” versus non-science?
- how could a line be drawn between the two?
- falsifiability: theory must be refutable by observation
- must be shown to be wrong by evidence
However:
- falsifiability is too easy, too open
- matter of consensus
What theories are there on the relationship between our mind and brain
1) Monism:
- only the mind exists
- only matter
- scientists view typically
2) Substance (dualism)
- mind and brain distinct
- don’t interact
Monism 1. materialism and 2. functionalism view
- mind as matter, only behaviour
2. mind as what brain does
What do psychologists do
- study various topics (e.g perception, memory, learning, language)
- 5 main perspectives:
1) biological
2) cognitive
3) differential
4) developmental
5) social - all based on observation and measurement
Biological perspective
- biological bases of the mind
Cognitive perspective
Cognition, overlaps with many areas/approaches
Differential perspective
Differences among individuals, how and why we differ e.g in intelligence
Developmental perspective
- how we change over time, infants to old age, how changes occur, how we learn etc
Social perspective
How we interact with eachother, how we form groups, how we judge other groups etc
How can we observe the mind
- cannot be directly observed
- depends on observable responses to things
- there is a gap between what we study (mind) and what we observe (responses)
How can we measure the mind
- need methods that provide quantitative responses (numerical data) relating to these
- we cannot observe, then measure, we observe via measurement
What does measurement usually require
- a zero point
(e. g starting area when measuring a room) - unit of measurement
(e. g measuring tape)
Unit of measurement for thought
- just noticeable difference between stimuli = smallest difference that you can detect
Zero point for thought
- the absolute threshold: smallest detectable stimulus
How is a mathematical relationship with physical world and psychological world possible
All that is needed is a zero point and unit of measurement
- relationship between physical (e.g weight) and mental (e.g detection of weight, difference in weight)
Mental chronometry is an example of what
- measuring the mind
1) measures ability to distinguish between stimuli
2) observes time it takes/ chronometer shows the time
What does Ebbighaus think about memory and how you could study memory in a scientific way
- recall depends on familiarity, meaning
- created a list of syllables (e.g tax, bok
- tested ability to recall list of syllables until memorised
- measures the performance on a recall task involving meaningless syllables
- observes a very specific and artificial form of remembering
Differences in what we study, observe and measure
What we study - the ‘stuff’ and phenomena we think and feel which makes ur who we are e.g memory, happiness
What we observe- psychological constructs, concepts with labels, e.g STM, IQ
Constructs are linked to measurements given to people. They respond in ways which is countable. This makes it quantitative, gives us data.
What we measure- psychological measurements used (tests, scales) to get data from sample
What is key to remember about what we study, observe and measure
- there is a gap between what is studied, observes and measured
Theoretical construct
- ‘Things’ in psychology that we are interested in, however have no access to e.g personality
Step of design/measurement
- must occur= heart of operationalisation e.g experiment
- this will allow you to find a piece of information as close to the original construct as possible
- this is the data
Relationship between constructs and variable
Ideally this relationship is right and as close as possible
Operationalisation
- process of defining the measurement of a phenomenon that is not directly measurable
- to represent the decisions we make about how we are going to go about measuring the thing we are interested in
Typically we talk about data having different levels of measurements
The typical categorisation
- nominal: numeric values indicate group
- ordinal: numeric values order responses
- interval: numeric values with no meaningful 0
- ratio: numeric values with meaningful 0
Operationalisation image process
CONSTRUCT A theoretical
⬇️
MEASURE A measurement or design
⬇️
VARIABLE A data
Representational vs pragmatic
Representational - something we can clearly see/touch, has measurement of clearly defined scale, relationship between thing and scale is easy to understand
Pragmatic - much more often that the thing has no independence of the thing used to measure it.
Difficult to say the thing is dependable from the measure
Example of representational measurement
- typical measurement, what we all understand about measurement
- e.g height
Where do psychological measurements fit between representational and pragmatic ones
Sit somewhere in between
- IQ
- Brain volume as measured by brain scanner
Jingle and jangle
- sometimes measures of different things are given the same name (Jingle)
- sometimes measures of the same thing are given different names (Jangle)
What is reliability
Accuracy of our measurement:
- across items (internal consistency)
- across persons (inter-rater agreement)
- across time (test-retest)
Reliability:
How our measurement should be accurate across items and issues with this
- if measured have all been identical, items should correlate (e.g personality Qs very, not confident)
= strong relationship between items
= internally consistent
Issue:
If measurements are too similar this will improve reliability however we may only be measuring very narrow area of broader area
Reliability:
How our measurement should be accurate across persons and issues with this
- if someone has specific, distinct qualities/personality traits, other people should have roughly consistent knowledge and information of this.
= a good measure arguably has inter-rater consistency as responses are consistent
Issue:
-affected by context and experience
- can be disturbed if a person behaves differently with differently types of people or contexts e.g at home vs work, social event vs home with parents
= we may only be measuring some downward result of the theme
Reliability:
How our measurement should be accurate across time and issues with this
- correlation between results in a test at different points e.g people will get similar results when given the same test one day and again a week later
= assuming thing we are measuring is stable and same questionare, results will correlate
Issue:
- may be affected by any situational factors e.g mood, being tired, may affect how people respond
- if factors are connected to thing being measured can be quite big e.g anxiety questionnaire filled out at home vs watched
- may be affected also by time between measures
Validity
Concerns the degree to which our measurement tool measures what it was designed to measure
- tied to operationalisation
- a subjective impression about the content and structure of our measure vs our definition
Predictive validity
Does the measure relate to things which it should relate to
Types of validity traditionally
1) content validity:
- test/measure contains on content relevant to the intended construct
2) face validity:
- does it appear to measure the intended construct
Types of validity Cronbach and Meehl
1) convergent validity:
- high correlations with measures of the same construct
2) discriminant validity:
- low correlations with measures of different constructs
Typed of validity, associations over time
1) concurrent validity:
- the correlations above are derived from contemporaneous measures
2) predictive validity:
- the construct predicts an expected outcome at same time in the future
Measurement error
Measurement error is the difference between a measured value and the true value. It means accurate representation is not possible.
- measured value= data point we have for a person
- true value= someone’s actual inherit level of the thing
What is measurement error comprised of
Random error :
Inevitable, things are never perfect. Errors can only be minimised
Systematic error:
Will affect everybody, can be reduced with good study design and measure selection e.g using faulty measurement tool
Within measurement error and operationalisation what is signal to noise
- how large the thing we are trying to measure is (signal), relative to all the things that make that measurement hard (noise)
- too much noise/margain for error makes it not an accurate way to measure
- we must design things/ measure things in a way to minimise the noise that exists in our system. Therefore we can successfully be able to get estimates of the things we are interested in
Why and when is it important to minimise noise in our system when measuring/designing
To improve certainty in results when:
- constructs are only partially defined
- if we have vague, untestable relations between constructs and their operationalisations
- different degrees of reliability and validity in our measurement tools
- different amounts of measurement error
What do psychologists do
- study various topics (e.g perception, memory, learning, language)
- 5 main perspectives:
1) biological
2) cognitive
3) differential
4) developmental
5) social - all based on observation and measurement
Biological perspective
- biological bases of the mind
Cognitive perspective
Cognition, overlaps with many areas/approaches
Differential perspective
Differences among individuals, how and why we differ e.g in intelligence
Developmental perspective
- how we change over time, infants to old age, how changes occur, how we learn etc
Social perspective
How we interact with eachother, how we form groups, how we judge other groups etc
How can we observe the mind
- cannot be directly observed
- depends on observable responses to things
- there is a gap between what we study (mind) and what we observe (responses)
How can we measure the mind
- need methods that provide quantitative responses (numerical data) relating to these
- we cannot observe, then measure, we observe via measurement
What does measurement usually require
- a zero point
(e. g starting area when measuring a room) - unit of measurement
(e. g measuring tape)
Unit of measurement for thought
- just noticeable difference between stimuli = smallest difference that you can detect
Zero point for thought
- the absolute threshold: smallest detectable stimulus
How is a mathematical relationship with physical world and psychological world possible
All that is needed is a zero point and unit of measurement
- relationship between physical (e.g weight) and mental (e.g detection of weight, difference in weight)
Mental chronometry is an example of what
- measuring the mind
1) measures ability to distinguish between stimuli
2) observes time it takes/ chronometer shows the time
What does Ebbighaus think about memory and how you could study memory in a scientific way
- recall depends on familiarity, meaning
- created a list of syllables (e.g tax, bok
- tested ability to recall list of syllables until memorised
- measures the performance on a recall task involving meaningless syllables
- observes a very specific and artificial form of remembering
Differences in what we study, observe and measure
What we study - the ‘stuff’ and phenomena we think and feel which makes ur who we are e.g memory, happiness
What we observe- psychological constructs, concepts with labels, e.g STM, IQ
Constructs are linked to measurements given to people. They respond in ways which is countable. This makes it quantitative, gives us data.
What we measure- psychological measurements used (tests, scales) to get data from sample
What is key to remember about what we study, observe and measure
- there is a gap between what is studied, observes and measured
Theoretical construct
- ‘Things’ in psychology that we are interested in, however have no access to e.g personality
Step of design/measurement
- must occur= heart of operationalisation e.g experiment
- this will allow you to find a piece of information as close to the original construct as possible
- this is the data
Relationship between constructs and variable
Ideally this relationship is right and as close as possible
Operationalisation
- process of defining the measurement of a phenomenon that is not directly measurable
- to represent the decisions we make about how we are going to go about measuring the thing we are interested in
Typically we talk about data having different levels of measurements
The typical categorisation
- nominal: numeric values indicate group
- ordinal: numeric values order responses
- interval: numeric values with no meaningful 0
- ratio: numeric values with meaningful 0
Operationalisation image process
CONSTRUCT A theoretical
⬇️
MEASURE A measurement or design
⬇️
VARIABLE A data
Representational vs pragmatic
Representational - something we can clearly see/touch, has measurement of clearly defined scale, relationship between thing and scale is easy to understand
Pragmatic - much more often that the thing has no independence of the thing used to measure it.
Difficult to say the thing is dependable from the measure
Example of representational measurement
- typical measurement, what we all understand about measurement
- e.g height
Where do psychological measurements fit between representational and pragmatic ones
Sit somewhere in between
- IQ
- Brain volume as measured by brain scanner
Jingle and jangle
- sometimes measures of different things are given the same name (Jingle)
- sometimes measures of the same thing are given different names (Jangle)
What is reliability
Accuracy of our measurement:
- across items (internal consistency)
- across persons (inter-rater agreement)
- across time (test-retest)
Reliability:
How our measurement should be accurate across items and issues with this
- if measured have all been identical, items should correlate (e.g personality Qs very, not confident)
= strong relationship between items
= internally consistent
Issue:
If measurements are too similar this will improve reliability however we may only be measuring very narrow area of broader area
Reliability:
How our measurement should be accurate across persons and issues with this
- if someone has specific, distinct qualities/personality traits, other people should have roughly consistent knowledge and information of this.
= a good measure arguably has inter-rater consistency as responses are consistent
Issue:
-affected by context and experience
- can be disturbed if a person behaves differently with differently types of people or contexts e.g at home vs work, social event vs home with parents
= we may only be measuring some downward result of the theme
Reliability:
How our measurement should be accurate across time and issues with this
- correlation between results in a test at different points e.g people will get similar results when given the same test one day and again a week later
= assuming thing we are measuring is stable and same questionare, results will correlate
Issue:
- may be affected by any situational factors e.g mood, being tired, may affect how people respond
- if factors are connected to thing being measured can be quite big e.g anxiety questionnaire filled out at home vs watched
- may be affected also by time between measures
Validity
Concerns the degree to which our measurement tool measures what it was designed to measure
- tied to operationalisation
- a subjective impression about the content and structure of our measure vs our definition
Predictive validity
Does the measure relate to things which it should relate to
