Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
62 Cards in this Set
- Front
- Back
Type I Error
|
Reject the Null, when really there was no difference between them, so it should not have been rejected.
|
|
Type II Error
|
Fail to reject the Null, when infact there was a difference and it should have been rejected.
|
|
Standard Deviation
|
How reliable is the central tendency/mean? Average amount by which scores deviate from the mean.
|
|
Variance
|
Used as part of the sum for SD, but rarely reported as its' units are squared
|
|
Confidence Intervals
|
Tell us whether we can be 95% confident that the interval which has been calculated captures the population mean
|
|
Effect Size
|
Provides an estimate of the magnitude of difference among sets of scores, while at the same time taking into account the amount of variability in the scores
|
|
Power
|
Is the probability of detecting 'actual' differences.
|
|
Intersection / Partially Dependent
|
A and B / A n B (when a part of each is overlapping)
|
|
Union
|
A or B / A u B.
|
|
Mutually Exclusive
|
There is no intersection
|
|
Mutually Exclusive and Exhaustive
|
No Intersection, and there is also no space, or possibility for anything else to happen. A + B + C = Complete Set
|
|
p (BlA) = p (AnB) / p (A)
|
This means the probability of B given A = probability of A & B / probability of A
|
|
Bayes Theorem
|
Calculates the probability of your data, not the probability of Ho. Bayes reverses the conditional probability equation. So now, p (AlB) = p (A) p(BlA) / p(B)
|
|
p(AlB) = p(A) p(BlA) / p(B)
|
This means the probability of A given B = probability of A x probability of B given A / probability of B
|
|
Belief Updating
|
B = observed state of the world
p(A), p(B) are 'priors' p(BlA) is a know relationship p(AlB) is the 'posterior calculation' |
|
Unordered Combinations
|
'Choosing' r objects from n total objects when you no longer care about the order in which they are chosen.
|
|
Binominal Cooefficent
|
(N) = N! / r! (N-r)!
(r) e.g what is the probability of getting HTTHH coinflips? 5!/3! 2! = 5 x 4 x 3 x 2 x 1 / 3 x 2 x 1 x 2 x 1 = 5 x 4 / 2 x 1 = 20 /2 = 10. |
|
Range
|
(largest minus smallest)
pro: easy to calculate con: sensitive to sample size |
|
Mode
|
(most common)
pro: best guess if you need to be exactly right con: may require 'binning' data |
|
Median
|
(middle value)
pro:insensitive to outliers con:no good inferential stats |
|
Mean
|
(average)
pro: central limit theorem con: highly sensitive to outliers |
|
Central Limit Theorem
|
When there is a sufficiently large sample, there will be a normal distribution
|
|
A Priori Method
|
Knowlege w/o experience
Using reason to develop a consensus among those debating the merits of one belief over another - An unmarried man is a bachelor - The pope is an unmarried man - Therefore, the pope is a bachelor |
|
Empiricism
|
The process of learning things through direct observation or experience
|
|
Belief Perseverance
|
A tendency to hold onto a belief even in the face of evidence which would convince most people the belief is false
|
|
Confirmation Bias
|
A tendency to search out info that supports one's beliefs whilst ignoring contrary info.
|
|
Availability Heuristic
|
When we experience an unusual or very memorable event, and overestimate how often such events typically occur
|
|
Objectivity
|
Eliminating such human factors as expectation and bias
|
|
Empirical Questions
|
Those that can be answered through systematic observations & techniques that characterize scientific methodology
|
|
Effort Justification
|
Basically cognitive dissonance
|
|
Reliable
|
Basically repeatablitly and a lack of errors
|
|
Validity
|
Have it measured what it was suppose to measure
|
|
Content Validity
|
Does the content 'make sense' in terms of the construct being measured
|
|
Face Validity
|
Whether the measure seems to be valid to those taking the test
|
|
Criterion Validity
|
Whether it can
(a) accurately forecast future behavior (b) is meaningfully related to some other measure of behav. |
|
Construct Validity
|
Whether it adequately measures the construct
|
|
Convergent Validity
|
Should be related to scores on other tests of the same construct
|
|
Discriminant Validity
|
Not related to scores on other tests which are unrelated to the construct
|
|
Nominal Scale
|
Use when asked empirical Q's
Use when assign to one group or another Use for frequencies, % etc Often use Chi-Squared Test |
|
Ordinal Scale
|
For rankings/relative position
e.g. hottest day, 2nd hottest Use stats like gamma correlation |
|
Interval Scale
|
Rankings with equal intervals
e.g. degrees fahrenheit or celsius or IQ Use stats like T-tests or ANOVAs |
|
Ratio Scale
|
Like ordinal or interval but with a true zero point
e.g. degrees Kelvin Use stats like t-tests or anovas |
|
Systematic Variance
|
Result of some identifiable factor, either the variable of interest or a confound.
|
|
Error Variance
|
Non-systematic variability due to individual differences or random unpredictable events
|
|
Standard Scores (Z scores)
|
Measure of how differenct some value is
How many SD's from the mean Mean of Z is always 0 SD of Z is always 1 Why do it? All values are placed on a common scale, so you can find and eliminate outliers |
|
Basic Research
|
Concerns describing/predicting/explaining fundamental principles of behavior.
|
|
Applied Research
|
Has direct & immediate relevance to the solution of a real-world problem
|
|
Mundane Realism
|
How closely a study mirrors real-life experiences
|
|
Quantitative Research
|
Numbers basically
|
|
Qualitative Research
|
Detailed info not in number form
|
|
Operationism
|
Science must be totally objective & precise, all concepts should be defined in terms of a set of operations to be performed
|
|
Operational Definitions
|
e.g. the length of some object could be defined operationally by a series of agreed upon procedures
|
|
The Zeigarnik Effect
|
A phenomenom where recall is better during an incomplete task, rather than completed ones
|
|
Logical Fallacy of Affirming the Consequent
|
-If the bird is a crow, then it will be black
-Here's a black bird -Therefore it must be a crow (Fallacy as not all black birds are crows) |
|
Logically Correct Modustollens
|
-If the bird is a crow, then it will be black
- Here is a yellow bird - Therefore, it cannot be a crow (correct as it has been asserted that all crows are black) |
|
Productivity
|
Good theories advance knowledge by generating a great deal of research
|
|
Good Theories are 'Parsimonious'
|
Parsimonious basically means simple, so the more simple the theory the better it is
|
|
Converging Operations
|
Different operational definitions producing similar results
|
|
Phonological Loop
|
-'a 2 second loop of recorded speech'
-memory span is actually less than 7 for longer words -explains why there are confusions based on sound |
|
Proactive Inhibition
|
Where the prior existence of old memories makes it harder to recall/learn newer memories.
|
|
Retroactive Inhibition
|
Where material learned later disrupts retrieval of info learned earlier.
|
|
Release from Proactive Inhibition
|
When you find it hard to remember new memories due to them being similar to previously learned ones, and you release from this as you move to a different kind of stimulus
|