Scientific Methods & Descriptive Statistics Flashcards
Define:
What is a standard definition for a p-value?
A p-value represents the probability of obtaining a test statistic as extreme as the one observed from our dataset given the assumption of the null hypothesis being true
In other words, if there is an extremely low probability of getting the result that you did under these conditions, then you have cause to reject the null hypothesis (since it is so unlikely to be true and for you to have gotten that particular test statistic)
List:
Give four key ‘goals’ of psychological scientific research:
(‘Goals’ = potential post-research applications)
- Describing
- Predicting
- Explaining
- Controlling
(Human behaviour)
Define:
Psychological Science
“Discovering how and why people think, feel and behave [in certain ways] through systematic accumulation of evidence”
(Source: Proffessor Matt Hammond, VUW, Lecture 1 Slides)
Compare & Contrast:
Inductive vs. Deductive research methods
- Inductive research begins with collation of data which you then derive patterns/generalisations from in order to contribute to new/existing theories
- Deductive research concerns verifying/building upon pre-existing theories and testing hypotheses (which are generated before collecting data)
Both contribute to the ‘systematic accumulation of evidence’ as according to general models of scientific research
List the missing labels for the ‘Model of Psychological Science’ pictured below
Note: This model also applies to other scientific research fields
- Theory
- Hypothesis
- Data
- Pattern Identification & Generalisation
A) Inductive Process
B) Deductive Process
Note that both inductive & deductive processes contribute to the scientific method, and this cyclical model allows for continuous systematic accumulation and refinement of evidence/knowledge
According to Karl Popper, what TWO key things constitute a ‘good theory’?
A ‘good theory’ is testable and falsifiable
- Testability = ability to be measured/assessed empirically
- Falsifiability = ability to be refuted/’proven wrong’
Define:
Theory
Theories are explanations for observed phenomena that are informed by scientific methods including carefully tested hypotheses and other solid data / evidence
Define:
A method
(In the context of scientific research)
The process(es) by which we go about testing predictions / hypotheses
Describe:
Based on the model below, the kind of data you cannot generate theories from?
‘Exceptional events’
These are impossible to derive meaningful patterns or generalisations from, which are the ‘bridging’ step between ‘data’ and ‘theory’
Identify:
FOUR key types of research designs
(According to the PSYC232 Course)
- Questionaire/Survey
- Naturalistic Observation
- Experimental Design
- Case Study
Identify:
Case studies are unique because they tend to…
- Involve small ‘N’s (i.e. an individual, group, community, etc.)
- Be more inductive
- Result in data that is too unique for any initial generalisation/broader-spanning pattern recognition
True or False:
We can prove our alternative hypothesis to be true if our data returns a statistically significant p-value
False
Observing statistical significance in the data gives us cause to reject the null hypothesis, but can never prove the alternative hypothesis (since we are finding evidence against the null hypothesis and not for the alternative)
Fill-in-the-Blank:
Both ____ and ____ scales have equal distances between each of their points of measurement.
However, ____ scales have no meaningful zero point and cannot use phrases such as ‘double’ or ‘half’ to compare its measurement points. A ____ scale though does have a ‘true zero’ (the absence of ‘x’) and can have measurements compared in those ways.
Both interval and ratio scales have equal distances between each of their points of measurement.
However, interval scales have no meaningful zero point and cannot use phrases such as ‘double’ or ‘half’ to compare its measurement points. A ratio scale though does have a ‘true zero’ (the absence of ‘x’) and can have measurements compared in those ways.
Interval scales can instead describe points as ‘more/less’ than another.
Both scales may use means for statistical analysis
Define:
Standard Deviation
‘Standard Deviation’ is used to describe the extent to which individual data points vary from the mean (of the population ‘N’, or the data-set/sample ‘n’ )
Define:
What is the central limit theorem?
The central limit theorem states that if you have a population with mean μ and standard deviation σ and take sufficiently large random samples from the population with replacement , then the distribution of the sample means will be approximately normally distributed
(Source: Boston University)
Note: the general minimum value of ‘n’ for this to apply is n = 30
Epistemology is concerned with what concepts?
Epistemology involves theories about knowledge itself, such as whether it is valid, or the methods we utilise in the search for it
What do we know? How do we know it? And do we truly know it?
Explain:
Why are means/averages not always entirely adequate to describe humans?
No one is perfectly average, and so variability is also vital when describing humans and their behaviour.
Think of the need to have adjustable seats - (you would struggle to find someone who fits the population average for every single bodily dimension these things are designed around!)
Define:
Operationalisation
This is the process by which researchers specify how they quantify/measure particular constructs
(e.g. quantifying ‘stress’ by indicators of physical arousal such as increased heart-rate, skin-conductance, etc.)
Fill-in-the-Blank:
Interval and ratio scales are used for ____ variables
continuous
Fill-in-the-Blank:
____ data increases/decreases in an order without mathematically meaningful distances between each point
Ordinal
(aka ‘Ranked’ data)
Because the extent/magnitude of difference between each piece of data is not clear, we tend not to use it as much in Psychological scientific research
Compare & Contrast:
Reliability & Validity
Reliability refers to how consistently you achieve a certain outcome/result, whereas validity questions whether you have appropriately measured something
Define:
What is item reliability?
Item reliability concerns whether or not each item of a scale measures the same thing
(i.e. internal consistency)
Define:
What is test-retest reliability?
Test-retest reliability concerns whether or not your measurements remain consistant across multiple repeats
(This can be improved by measuring things in the same manner, with the same equipment each time, etc.)
Define:
What is observer/rater reliability?
Observer/rater reliability concerns whether or not people make consistent ‘ratings’/’value judgements’ etc. with each other
(This can be decreased significantly if ambiguity in how one may measure something is increased.
Think of Amazon reviews where one person rates a product highly because it was the quality they were expecting, but another person rates it poorly because it arrived slowly - these two people are giving ratings with entirely different sets of standards in mind to each other)
Define:
What question embodies the concept of construct validity?
‘Do our measurements make sense with the initial theory?’
Define:
Internal validity concerns…
…whether or not our study design was the most appropriate for addressing the research question
Define:
External validity concerns…
…whether or not our findings coincide appropriately with other findings and concepts in the real world beyond our lab setting
Define:
Statistical validity concerns…
…whether or not the statistical test(s) we have chosen are appropriate for the original research question and measurement(s)
