Docsity
Docsity
Prepare for your exams
Prepare for your exams

Study with the several resources on Docsity

Earn points to download
Earn points to download

Earn points by helping other students or get them with a premium plan

Guidelines and tips
Guidelines and tips
Sell on Docsity
Sell on Docsity
Docsity AI
Docsity AI
Log inSign up

Prepare for your exams

Study with the several resources on Docsity

Find documents

Prepare for your exams with the study notes shared by other students like you on Docsity

Search for your university

Find the specific documents for your university's exams

Docsity AINEW

Summarize your documents, ask them questions, convert them into quizzes and concept maps

Explore questions

Clear up your doubts by reading the answers to questions asked by your fellow students

Earn points to download

Earn points by helping other students or get them with a premium plan

Share documents
download point icon20 Points

For each uploaded document

Answer questions
download point icon5 Points

For each given answer (max 1 per day)

All the ways to get free points
Get points immediately

Choose a premium plan with all the points you need

Study Opportunities

Choose your next study program

Get in touch with the best universities in the world. Search through thousands of universities and official partners

Community

Ask the community

Ask the community for help and clear up your study doubts

Free resources

Our save-the-student-ebooks!

Download our free guides on studying techniques, anxiety management strategies, and thesis advice from Docsity tutors

PSYC 2080 MIDTERM 1 VERIFIED ACCURATE STUDY GUIDE, Exams of Psychology

United States International UniversityPsychology

PSYC 2080 MIDTERM 1 VERIFIED ACCURATE STUDY GUIDE

Typology: Exams

2025/2026

Available from 04/04/2026

ROCKY-B
ROCKY-B 🇰🇪

4.4

(16)

40K documents

1 / 15

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
PSYC 2080 MIDTERM 1 VERIFIED ACCURATE STUDY
GUIDE
Measurement - Answers - act of identifying properties of an object.
Measurement Theory - Answers - field of stats that describes and evaluates quality of
measurements
latent construct - Answers - latent- unobservable
construct- domain of behaviours
operational definition - Answers - ex ability to recall digits is the ___ for working memory
achievement test - Answers - assess prior learning
aptitude test - Answers - assess potential to learn, specific task/skill/narrow
intelligence test - Answers - ability to solve problems and think abstractly/relative ability
in global areas
personality tests - Answers - assess traits, qualities, behaviours that determine
individuality (overt and covert)
Personality test (structured) - Answers - accept or reject statements of one's self, self-
report
Personality test (projective) - Answers - reactions to ambiguous stimuli are recorded
and interpreted
-Reveal aspects of unconscious mind
-Assumes responses reflect individual characteristics
-Ex inkblot tests
Neuropsychological test - Answers - assess cognitive, sensory, perceptual, and motor
performance to determine brain damage
Behavioural procedure - Answers - objectively describe and count frequency of
behaviour, identifying antecedents and consequences of behaviour
Interest inventories - Answers - measure preference for certain activities or topics and
help occupational choice
Creativity tests - Answers - assess novel, original thinking and capacity to find unusual
or unexpected solutions to problems
pf3
pf4
pf5
pf8
pf9
pfa
pfd
pfe
pff

Partial preview of the text

Download PSYC 2080 MIDTERM 1 VERIFIED ACCURATE STUDY GUIDE and more Exams Psychology in PDF only on Docsity!

PSYC 2080 MIDTERM 1 VERIFIED ACCURATE STUDY

GUIDE

Measurement - Answers - act of identifying properties of an object. Measurement Theory - Answers - field of stats that describes and evaluates quality of measurements latent construct - Answers - latent- unobservable construct- domain of behaviours operational definition - Answers - ex ability to recall digits is the ___ for working memory achievement test - Answers - assess prior learning aptitude test - Answers - assess potential to learn, specific task/skill/narrow intelligence test - Answers - ability to solve problems and think abstractly/relative ability in global areas personality tests - Answers - assess traits, qualities, behaviours that determine individuality (overt and covert) Personality test (structured) - Answers - accept or reject statements of one's self, self- report Personality test (projective) - Answers - reactions to ambiguous stimuli are recorded and interpreted

  • Reveal aspects of unconscious mind
  • Assumes responses reflect individual characteristics
  • Ex inkblot tests Neuropsychological test - Answers - assess cognitive, sensory, perceptual, and motor performance to determine brain damage Behavioural procedure - Answers - objectively describe and count frequency of behaviour, identifying antecedents and consequences of behaviour Interest inventories - Answers - measure preference for certain activities or topics and help occupational choice Creativity tests - Answers - assess novel, original thinking and capacity to find unusual or unexpected solutions to problems

Test - Answers - technique to quantify a behaviour, systematic procedure for comparing behaviour of two people (Cronbach) test item - Answers - specific stimuli that produces response, many items for proper test psychometrics - Answers - science of evaluating characteristics of psychological tests Criterion referenced tests - Answers - decisions made in comparison to a cut-off score ex min score Norm-referenced tests - Answers - score is compared to reference sample o expected score from population

  • test taker needs to be part of ref population Galton - Answers - father of psychometry Wundt - Answers - father or psychology as a science
  • large impact on creation of behavioural tests (ex Working Memory) 1905 Binet-Simon Scale - Answers - First intelligence test to create standardized sample of 50 children (Standard Condition) 1916 Stanford-Binet Intelligence scale - Answers - many items + increased standard sample to 1000 individuals Army Beta - Answers - An intelligence test developed during WW1 and used by army for soldiers who cannot read. Trait - Answers - relatively enduring dispositions that distinguish one person from another Descriptive Statistics - Answers - summarize information in a collection of data Inferential Statistics - Answers - provide predictions about a population based on data from a sample of that population Parameter - Answers - summary information for the population 4 major variable types - Answers - ratio, interval, ordinal, nominal Ratio data - Answers - - continuous
  • can rank values
  • Zero is meaningful
  • addition, subtraction, multiplication, division
  • ex income, age, time

u= population average N= population number

  • most common measure of dispersion
  • (sqrt of variance is SD) Poisson distributions - Answers - - long positive skewness
  • mean equal to variance standard normal distribution - Answers - A normal distribution with a mean (mu) of 0 and SD (sigma) of 1.
  • symmeterical Standard normal statistics/cumulative probability - Answers - where we are in probability distribution of Standard Normal Distribution
  • What is the probability that Z is less than our calculated value of Z* percentile - Answers - a value where a specific percent of remaining values will be less than it
  • Ex: top 1% means you are the 99th percentile and 99% of class is lower than you Central Limit Theorem (CLT) - Answers - for a given distribution, with sufficiently large n (n>30), we can use standard normal distribution and Z statistic to calculate probability Correlation - Answers - A measure of the relationship between two variables Properties of Correlation (r) - Answers - 1. The value of r does not depend of which variable we designate X or Y
  1. The value of r is independent of scale of X and Y
    • 1 <= r <= 1
  3. r=1 only if all (xi,yi) pairs lie on a straight line with positive slope. r=-1 in the case for a negative slope. Spearman's rho - Answers - correlation for ordinal categorical data Polychoric and tetrachoric Correlation - Answers - correlation of 2 ordinal variables which rests upon an assumption of an underlying joint continuous distribution When the two variables are dichotomous it is called a Tetrachoric correlation Reliability - Answers - the degree to which scores are free of error
  • a measure of consistency
  • cant be valid without being reliable True Score Model for classical test theory (CTT) - Answers - x=t+e x= observed fallible score t=true score e= error score
  • Charles spearmen 1904 random variable - Answers - values vary and are random because they depend on specific characteristics sample space - Answers - the set of all possible outcomes ex sample space of course is either pass or fail (0/1) Assumption 1 of True Score Model - Answers - 1. The attribute that the test is measuring remains constant
  • errors of test are independent and randomly distributed
  • means a score will remain constant and any differences are sample-specific and random error Assumption 2 of True Score Model - Answers - - true score for a person = mean of their observed scores over infinite trials
  • At group-level, true score = mean of infinite number of individuals writing test Assumption 3 of True Score Model - Answers - observed score variance= true score variance + error score variance
  • correlation btwn true score and error = 0
  • assumption of independence
  • not an error process that differs among individuals Assumption 4 of True Score Model - Answers - - When independent sample of persons take two separate tests parallel in structure, correlation btwn error of two tests = 0
  • means two random variables (test 1 & 2) are truly random and independent
  • Without this, we cannot compare two tests as the difference in scores may be a consequence of different error processes Assumption 5 of True Score Model - Answers - - Error scores on one test are uncorrelated w true scores of another test

observed SD * root(1-coefficient of reliability) domain sampling theory - Answers - · Avoid making assumption of parallel tests · have a limited number of items out of larger pool of items · assume subset represents larger pool of all items · Then two tests can be constructed out of diff subsets of items and correlation between them would estimate reliability · Reliability conceptualized by how well true score is measured when using asubset of all items · Ex Depression, we assume that a given depression test could have been constructed with diff items from pool of depression items generalizability theory - Answers - · extension of CTT · Can deal with fact that a test may measure multiple concepts rather than assumed single · helps relax many assumptions by introducing a more complex model to estimate reliability Item Response Theory - Answers - · Solves several problems with CTT · Ex: under IRT individuals can receive different versions of test (Under CTT, test composition is integral to reliability) · IRT expanded w advancement of computers commonly used in field of education test-retest - Answers - · The same test applied on two different occasions · can examine error associated w administering test at diff times · correlation between tests is reliability (coefficient of stability) · only meaningful if true score does not change over time · must also assume error variance is stable over time carryover effects - Answers - where information gained by writing test carries over to the second time writing · When carryover effect is consistent, no impact to reliability · carryover only shifts response distribution left/right and does not change reliability · becomes impossible to estimate true reliability because we do not know direction of bias (reliability over- or underestimated?) · acceptable reliability of method ranges from >0.9 (education) to 0.8-0.9 (personality) parallel forms reliability - Answers - · Compare two equivalent forms of test that measure same construct but made of diff items from pool of items, given at same time · link to domain sampling theory · coefficient of equivalence · When two tests given at same time, only diff in variability should be test items

· not used often due to six assumptions · sometimes no resources so parallel form would be 2 tests constructed out of single test post-administration internal consistency - Answers - assessment of reliability via within-test configurations split-half reliability - Answers - - Divide a test into two halves, scored separately, and compared

  • split should ensure Mean Observed Score and Error variance are equal to meet parallel forms assumption
  • correlation btwn scores of splits should equal coefficient of equivalence (not the case)
  • correlation btwn two halves will underestimate reliability · need to correct for this using Spearman-Brown Formula Spearman-Brown formula - Answers - · adjusts estimated reliability based on total # of test items · formula assumes equality of variance - an alternative called Rulon's formula relaxes this assumption: · Ex: CES-D was split in half and a correlation between two halves was 0.78. True reliability would be: 2(0.78)/1+0.78 = 1.56/1.78 = 0. rt= split half reliability based on SB formular rh=correlation btwn half tests Item homogeneity - Answers - · means all items are equal in importance and difficulty · If items differ in difficulty then split should be an equal balance between two halves · split-half process depends on having item homogeneity Cronbach's/ coefficient Alpha - Answers - · alternative to avoid test splitting or meeting CTT assumptions · most common measure of reliability, used for any item format · When tests are tau-equivalent or congeneric, we can use Cronbach's alpha to estimate reliability
  • estimating reliability summed across N-1 tests, where N is # of items in measure · estimates lowest possible reliability · when alpha is high (0.90), true reliability is between 0.90-0. · when alpha is low (0.60), true reliability is between 0.60-0. · confirms if a test is reliable but does not indicate if unreliable KR20 - Answers - · formula to calculate reliability that does not depend on a specific split or having to score a split separately · Only dichotomous items (0/1 scoring) · calculates variance of item scores as a proportion of total test variance · adjusts that tests with less items have more error variance

highly, low - Answers - If two tests are ___ correlated, reliability of difference score will be ___ change score - Answers - When difference score is between same test discrepancy score - Answers - When difference score is between two different tests more - Answers - change score is ___ likely to be correlated than discrepancy score Validity - Answers - whether a test measures what it is meant to measure Guidelines of validity - Answers - · 1. The test response process · 2. The internal structure of the test · 3. The relationship between the test and related variables · 4. The consequence of testing construct validity - Answers - does test appear to mean what we intend it to? (Whether items are mathematically related to construct) · In addition to item correlations we ask how test relates to other types of tests · When no good criterion exists, construct validity becomes even more important · when construct not well defined, construct validity can help operationalize construct by seeing how different items relate to it · For multidimensional tests, may need multiple other tests to establish differing relationships · Cronbach was supporter of establishing construct validity model · Some argue Construct Validity is backbone that holds all other validity types together Correlational, Group differences, Factor analytical, and Multitrait-multi-methods - Answers - Four major studies to examine construct validity Substantive Stage, Structural Stage, Final state - Answers - 3 stages of construct validity Substantive Stage of construct validity - Answers - · We define theoretical domains of construct · How should construct be measured? · This stage is theory driven · Strongly linked to content validity structural stage of construct validity - Answers - · What are the internal relationships among items and domains that operationalize the construct? · Do items behave in ways that we would expect based on our theories and research? · examine how items relate to latent construct (item loading)

· also examine for possibility of subscales, subsets of items that are very correlated to each other but not to others. · use Exploratory and Confirmatory Factor Analysis (EFA, CFA) to establish this stage · look to obtain moderate to high item correlations and factor loadings and for the internal structures to reflect theory rather than being a fragment from statistical method · Item correlations help assess item homogeneity - items of a construct should relate to a common trait/attribute · want homogeneous tests because they have better psychometric properties (ex reliability is higher when items are homogenous) · Item homogeneity is typically first thing examined in structural stage · No criteria for how high a correlation needs to be for items to be considered homogenous item loading - Answers - how items relate to latent construct subscales - Answers - subsets of items that are very correlated to each other but not to the others. item homogeneity - Answers - items of a construct should relate to a common trait/attribute

  • want homogeneous tests because they have better psychometric properties (ex reliability is higher when items are homogenous)
  • Item homogeneity typically first thing examined in structural stage
  • No criteria for how high a correlation needs to be for items to be considered homogenous final stage of construct validity - Answers - · examine external relationships among test (and our construct) with other related and unrelated constructs (Convergent and Divergent/Discriminant Validity) · When possible, also examine test with different groups · Related to Criterion validity when a criterion is what differentiates individuals · Ex: A psychiatrist designates individuals with clinical depression, subclinical depression, or not depressed. The psychiatrist designation is the criterion. As part of construct validity we may ask "can our test identify each of the groups". In this example we would provide evidence towards both Construct and Criterion validity convergent validity - Answers - test positively correlates with tests with similar constructs / items Divergent/Discriminant Validity - Answers - test negatively correlates with tests with dissimilar constructs / items and does not correlate at all with ones completely unrelated. criterion validity - Answers - does test correspond with intended criterions?
  • commonly calculated with Flesch-Kincaid score Flesch-Kincaid score - Answers - calculates readability 0.39 (total words / total sentences) + 11.8 (total syllabus / total words) - 15. underestimated, overestimate - Answers - If reliability is ___, correction will ___ correlation between the test and criterion factor analysis - Answers - · Technique to identify relationship among items to a hypothetical latent object called a Dimension. · We will designate this dimension as latent construct · Factor analysis looks to see what the min # of dimensions to account for all intercorrelations among test items · similar to other item reduction techniques such as Principal Component Analysis. · developed by Charles Spearman in 1904 during studies of intelligence · Spearman noted some variables from a domain are correlated with other variables from same domain, and so these must be sharing information about domain · posits that items correlate because they are determined by some common but unobserved influence (common factors) · unobserved influence is manifested as latent factor of interest. Exploratory Factor Analysis (EFA) - Answers - · Goal to identify set of factors that minimally represent item intercorrelations · Ex: I have 50 items that I believe represents Humour. Using EFA I can see if items seem to be related to a single dimension of humour, or several dimensions (different types of humours?), or a complex hierarchical structure with multiple latent constructs being related to a higher-order construct (types of humours related to an overall domain of humour). · can use EFA to identify how items relate to possible constructs · Items that correlate weakly to construct are not related · can help identify items that could be removed from test because they are not related to construct of interest / not related to the other items (impacting the validity and reliability). · just choose # of dimensions we believe exist, method presents us a pattern of how items would relate to our chosen number of dimensions · can then fine tune pattern through a process called Rotations How EFA works - Answers - · Give a test to a group of individuals; score test; input all item scores into a dataset, Feed raw scores into statistical software, run EFA or PCA on data · Program calculates Eigenvalue used to identify dimension # · software displays eigenvalues for each factor that could exist ( factor = construct) · One technique is select # of factors where Eigenvalue > 1 · if first factors have values of 4, 2, 0.87, dimensionality is 2 · A graphical alternative is called a Scree plot (plots Eigenvalues)

· When Screen plot sees a substantial drop and levels off, called the hinge (point of drop is dimensionality) · then rotate items: plot all items on graph, where each axis is a dimension · distance from axis indicates the factor loading, similar in concept to correlation or relation of item to a specific dimension · By rotating axis we can try to maximize relationships btwn items and dimension / construct. · Rotations necessary because factor loadings of item are not set; there is an infinite number of solutions to satisfy mathematical equation that is used to attach items to each dimension (factor indeterminacy) · means our goal is to rotate dimensions to identify a set of factor loadings that satisfy some equation while maximizing # of 0s (no loadings) on other dimensions while maximizing item correlations Orthogonal rotation - Answers - dimensions not allowed to be related to each other; forced to be independent by program

  • items can only relate to one dimension and not other Oblique rotation - Answers - · dimensions can correlate to each other (not guaranteed)
  • items will primarily load onto one dimension and secondarily load on others
  • important if dimensions relate to each other EFA - item loadings - Answers - · strength of relationship btwn an item and construct is estimated by item loading · stronger the loading, more related the item is to construct · Low loadings suggest items do not relate - can be considered for removing · No specific values established for what is strong or not but general idea is like a correlation Confirmatory Factor Analysis (CFA) - Answers - · Opposite of EFA - rather than letting program identify relationships, we specify them · program tells us how well our theoretical model fits data we have provided it · We must show every single relationship - item to construct, item to item, relationships between estimated errors · If model fit is bad we know we have mis-specified model in relation to our data · Let us test whether a test has suggested internal structure. Test Dimensionality - Answers - · Related to Construct Validity · How many latent constructs does test have? · Dimensionality is a notion of how many latent constructs exist in a test · unidimensional (1 construct) (homogeneity/conceptual homogeneity) multidimensional (2+ constructs)