SSR Exam 4 Flashcards
global network analysis
looks at the overall structure of the network
local network analysis
looks at the differences between nodes in terms of connectivity, centrality etc
nodes
set of entities
edges
nodes are connected through
hysterises
there is more needed for high connectivity networks (vulnerability network) to get back to normal when experiencing a traumatic life event
degrees
number of connections
betweenness
how often does a node lie on a path between two other nodes
network theory
mental disorders are alternative stable states in a symptom network
small world structure
high level of clustering , small average path lengths.
goal of psych assessment
to characterise an individual’s standing on individual differences construct of clinical relevance
latent trait models
posit the presence of one or more underlying continuous distributions
latent class models
based on the supposition of a latent group (class) structure for a personality construct’s dikstribution, and they are typically evaluated via latent class analysis
hybrid models
these models combine the continuous aspects of the latent trait models with the discrete aspects of latent class models
zones or rarity
locations along the dimension that are unoccupied by some individuals
quasi-continous
construct would be bounded at the low end by zero, a complete absence of the quality corresponding with the construct.
discrimination
measure of how strongly the item taps into the latent trait
assumption of conditional independence
classes are defined by patterns of item endorsement across individuals, assuming that inter-item correlations solely reflect class membership
self organisation
process where some form of overall order arises form local interactions between parts of an initially disordered system. the process can be spontaneous when sufficient energy is available, not needing control by any external agent.
bimodality
in exactly the same circumstances two stable states are possible
hysteresis
is the dependence of the state of a system on its history
finite mixture models
the distribution of data (length) is not described by one distribution (normal distribution) but a weighted sum of distributions
latent classes
both observed and latent variables are categorical
wilcoxon signed rank test
a non parametric aletnartive to the paired samples t-test. it assigns + or - signs to the difference between two repeated measures
reflective latent variable model
the attribute is seen as the common cause of observed scores: neuroticism causes worrying about things going wrong
formative latent variable model
observed scores define or determine the attribute
subjective probability
degree of conviction we have in a hypothesis.
P(H|D) is proportional to P(D|H) x P(H) says that
your posterior is proportional to the likelihood times the prior
likelihood
if you want to update your personal probability in a hypothesis, the likelihood tells you everything you need to know about the data. it captures all support for a hypothesis provided by the data
likelihood principle
the notion that all the information relevant to inference contained in data is provided by the likelihood
probability density distribution
if the dependent variable can be assumed to vary continuously, (that is the values do not come in steps)
p-value
the probability of rejecting the null, given the null is really true
bayes theorem
says that posterior is proportional to likelihood times prior
flat prior/uniform prior
if the std dev is infinite , you think all population values are equally likely
bayes factor
Bayes equivalent to null hypothesis testing or significance testing.
what is bayes statistics?
it is when you use probability to represent uncertainty in all parts of a statistical model/ flexible extension of maximum likelihood
bayesian data analysis is a method for figuring out unknown that requires three things
data, a generative model, priors (what ifs the model has before seeing the data)
P(X)
degree of belief that X is true.
probability
is a measure of the degree of belief or confidence one has in the truth of a proposition
bayesian program steps
find a way of assigning a number to a person’s degree of belief, show that a rational betting strategy must satisfy the rules of probability theory, adopt bayes’ rule as a general principle for how to learn from experience
p value
prob of the data of encountering a test statistic at least as extreme as the one observed, given that the null nhypothesis is true
confidence interval
an x% conf interval for a parameter 0 is an interval (L,U) generated by an algorithm that in repeated use has an X% chance to capture the true value of 0
classical definition of probability
proportion of occurrence when a particular experiment is repeated infinitely often (inner different circumstances)
statistical evidence
change in conviction (concerning a given hypothesis ) brought about by the data
advantages of bayes factor
provides a continuous degree of evidence without requiring an all or non decision, allows evidence to be monitored during collection, differentiates between the ‘data support H0’ (evidence for absence) and ‘the data are not informative’ (absence of evidence)
difference between dimensions and typologies
typology is like categories and dimensions is you can have different points on the scale that you can be
what does a combination of a typology and dimension look like?
like a hybrid model
median split
is one method for turning a continuous variable into a categorical one. Essentially, the idea is to find the median of the continuous variable. Any value below the median is put it the category “Low” and every value above it is labeled “High.”
complex systems
consisting of many sub systems interacting non linearly, like weather, chemical patterns, brains etc
CUSP model
consists of two stable regions and two thresholds where sudden changes occur. a linear model does not have these features.
local independence
the underlying assumption of latent variable models. The observed items are conditionally independent of each other given an individual score on the latent variable(s). This means that the latent variable explains why the observed items are related to another.
hubs and edges
hubs = the symtpoms, edges = the connections between these symptoms
degree
number of connections
highest betweenness
how often does a node lie on a path between two other nodes
the disease model of mental disorders
the problems that people that people encounter in life are ‘symptoms’ of a reasonably small set of underlying disorders that cause these symtpoms (lung tumor = shotness of breath, chest pain and coughing up blood)
critical slowing down
the theory that in cases where a system is close to a critical tipping point the recovery rate should decrease. It occurs because a system’s internal stabilizing forces become weaker near the point where they break and the system moves into a new regime. Thus critical slowing down is posited to exist at phase transitions, such as ecosystem collapse
prior
a probability distribution of an uncertain quantity is the prob dis that would express one’s beliefs about this quanitty before some evidence is taken into account.
likelihood
formed from the joint probability of a sample of data given a set of model parameter values. it is viewed and used as a function of the parameters given the data sample
posterior
of a random event or an uncertain proposition is the conditional probability that is assigned after the relevant evidence or background is taken into account.
marginal likelihood
in which some para§meter varibels have been marginalized
epistemology
refers to a branch of philosophy that is concerned with the theory of knowledge and that tries to answer questions about how we can know and what we can know
ontology
refers to the assumptions we make about the nature of being, existence or reality
positivism
holds that the relationship between the world (events objects and other phenomena) and ur sense of perception of the world is straightforward:
empiricism
holds that our knowledge o the world must arise from the collection and categorisation of our sense perceptions/observatios of the world
deductive reasoning
in research this means reasoning which begins with theories, which are refined into hypothesis, which are tested through observations of some sort, leads to a confirmation or rejection of the hypotheses
realism
assumes that a reality exists independent of the observer
small q qualitative research
research that uses qualitative tools and techniques but within a hypothetic-deducitve framework
big Q qualitative researhc
refers to the use of qualitative techniques within a qualitative paradigm which rejects notions of objective reality or universal truth and emphasises contextualised understandings
nomothetic research
which seek generalisable findings that uncover laws to explain objective phenomena
idiographic research
which seek to examine individual cases in detail to understand an outcome
phenomenological
methods focus on obtaining detailed description of experience as understood by those who have that experience in order to discern its essence
inductive reasoning
means reasoning that begins with data, which are examined n light of a study’s research question
critical realist outlook
assumes that while a reality exists independent of the observer, we cannot know that reality with certainty.
social constructionist prspective
adopts a critical stance towards the taken-for-granted ways in which we understand the world and ourselves, such as the assumptions that the categories we use to interpret the world correspond to ‘real’ \objectve’ entities
relativist
stance in which reality is seen as dependnt on the ways we come to know it
reflexivity
refers to acknowledgement by the researcher of the role played by their interpretative framework or speaking positiion (incl theoretical commitments, personal understandings and personal experiences) in creating their analytic account.
sensitivity to context
the research should make clear the context of theory and the understandings creatd by previous researchers using similar methods and or analysing similar topics
commitment
said to involve demonstrating prolonged engagement with the research topic
rigour
relates to the completeness of the data collection and analysis
coherence
refers to the quality of the reserch narrative, the ‘fit’ between the reserch question and the philosophical perspective adopted
impact and importance
relate to the theoretical practical and socio-cultural impact of the study
transparancy
entails detailing every aspect of the processes of data collection and analysis and disclosing/discussing all aspects of the research process
pluralistic analysis
exploring the value of applying different qualitative methods with different ontologis and epistemologies to a single data set
collective is required for long run relative freqyency
the probability of some property(q) occuring is then the proportion of events in the collective with proporty ‘q’
alpha level
longterm relative frequency of mistakenly rejecting hypothesis H1 if it si true, also known as typ 1 error rate
beta level
longterm relative frequency of mistakenly rejecting hypothesis H2 if it si true, also known as typ 2 error rate or 1 - power
hypothetical-deductive model
or method is a proposed description of scientific method. . According to it, scientific inquiry proceeds by formulating a hypothesis in a form that could conceivably be falsified by a test on observable data.
qualitative research
is a process of naturalistic inquiry that seeks in-depth understanding of social phenomena within their natural setting.
quantitative research
based on equity theory about relationship satisfaction
central assumption qualitative research
human behaviour is not a simple response to stimuli, but it depends on how the external world is understood
grounded theory
how do social processes impact participants? use if : you know little about the subject, if existing theories are not sufficient, when developing a theory, when interested n perception of the world in participants
interpretative phenomenological analysis
how does a person view this in this context? the goal is to get a detailed picture of how an individual gives meaning to events. crucial here is the role of the researcher: how does the researcher experience the world(and in this case the participant?)
discourse analysis
how are experiences constructed? social reality is constructed by the way we communicate. How do people construct different versions of their social world, and what benefits do they derive from this?
critical realism
a reality exists independent of the observer but we cannot know that reality with certainty
relativism
reality is dependent on the ways we come to know it
transparency & coherence
coherence between different components
commitment & rigour
completeness and perseverance of the researcher
impact & importance
practical and theoretical consequences of the research
sensitivity to context
beliefs and position of participants & researchers ; place in context of other research and theory
ranking the data
finding the lowest score on giving it a rank of 1, then finding the next highest score and giving it a rank 2 and so on.
five principles of network theory
complexity, symtpom component correspodence, direct causal connections, mental disorders follow network structure, hysteresis
discrete observed variables (correct/incorrect)& continuous latent variables (intelligence)
item respons models
discrete OV ((correct/incorrect), discrete LV(disorder)
latent class analysis
coninuous OV(sub test scores) & continuous LV (intelligence)
factor models
Continuous OV (test sub scores) & discrete LV (disorder)
mixture models
ethnography
be part of a group for a long time, observe interactions
focus group
group interview, with a focus on interaction between the group members
