foundations exam

Inductive approach

starts with an observation and generates an abstract theory or explanation from a social phenomenon (specific —> general)

When did experiments begin in the social sciences?

Late 1800’s

Hawthorne Effect

Researchers concluded that workers modified their behavior simply because they were aware of being watched, studied, and given special attention

Hawthorne Experiments

• ran between 1924 to 1932 at the Hawthorne Plant in Ciscero, Il. Fundamentally influenced management theories

• Three main experiments

  • Illumination studies (1924-1927): tested whether better or worse lighting changed worker out put

  • Work structure experiments (1927-132): relay assembled studies, bank wiring observations, women’s workers interviews

deductive approach

start with abstract theory for social phenomena, then test an empirical hypothesis derived from that theory (general —> specific)

what is a theory?

a general argument of how things work

most theories are what?

testable/falsifiable, generalizable, probabilistic

Hypothesis

a hypothesis is a logical, testable claim about the expected relationship between two concepts or variables

variable

a variable is anything that takes on different values, something that can be measured

independent variable (IV)

the hypothesized cause of variation in a dependent variable (the cause)

dependent variable

the hypothesized effect of variation in an independent variable (the outcome)

casual relationship

a change in one thing leads to some change in another

spuriousness

when two variables appear related but tare actually the result of a third variable

confounding variable

a third variable that makes two concepts appear linked even when they are not

factorial designs

involve the manipulation of multiple independent variables at the same time

What are conditions in an experiment?

Conditions reflect different levels of the independent variable, or different treatments

concept

an abstract idea that can be named, defined, and measured in some way

conceptualization

precisely defining abstract terms and concepts and turning them into specific variables

operationalization

identifying how a variable or concept will be manipulated or measured

Doob and Gross (1968)

• question: how does status of a person impact how others respond to their frustrating behavior?

• Hypothesis: people will act more aggressively in response to their frustrating behaviors of low - vs- high status other

• Independent variable: (vehicle) status, high status vehicle vs low status vehicle

• Dependent variable: honking

• Findings: 84% honked at low status, 50% at high status car, men honked more often

Cornell et al. (2007)

• example of 2 × 2 factorial design

• Manipulates both gender and parental status of fake job applicants

Willer et al. (2013)

• looked at the relationship between masculinity threat and testosterone levels on support for President Bush

• Foundation that men with naturally high basal testosterone levels felt more masculinity threat and had more support for President Bush

random assignment

helps to eliminate or reduce the possible impact of potential confounding variable

law of large numers

potential confounds will begin to balance out between groups as the size of the groups increase

between-subjects design

participants are randomly exposed to only one study condition

within-subjects design

participants are exposed to more than one of the study conditions

Strengths of between-subjects

• reduced risk of carry-over and demand effects

• offers a more conservative test of hypothesis

weaknesses of between-subjects

• requires larger sample size

• may reduce external validity

strengths of within-subjects

• more economical because participants provide data on all measures

• may provide better external validity

weaknesses of within-subjects

higher risk of carry-over and demand effects

mediating varibale

a variable that explains why or how an IV causes a change in a DV

moderating varible

a variable that explains for whom or under what conditions an IV and DV

validity

how “accurate” a study is

three types of validity

internal, external, construct

Internal validity

eliminates or reduced the impact of compound variables, less question in the IV causing change in the DV

external validity

how generalizable are the results of the study

representativeness

how representative of the population is the sample being studied?

realness

to what extent does the study reflect the “real” world"?

mundane realism

how similar are the participants and study situations are to those the researcher wants to generalize to

construct validity

have concepts been operationalized in ways that will capture them accurately?

face validity

subjective assessment of whether a manipulation or measurement appears to capture a concept

reliability

how consistent is your measurement of manipulation

lab experiments strengths

• Maximum control over environment and measurements

• Standardized procedures increase potential for replication and reliability

• Can create more elaborate procedures to test more abstract theories and complex process

• Can collect certain types of measures not possible or difficult in other design (e.g., biomarkers, behavior)

lab experiments weaknesses

• Samples are often less representative and smaller 

• Artificial environment does not necessarily reflect the “real” world

• may elicit stronger demand effects

survey experiments

an experiment embedded in a survey

• Lab experiments

• Most heavily controlled form of experimental design

• Often use non-random samples

• May involve the use of confederates

survey strengths

• Larger and more diverse samples that may be more representative of the population of interest 

• Can manipulate a significant number of independent variables (e.g., conjoint)

Asch conformity experiments

• Method: confederates and one participant, had to choose which line matched the one presented

• Had confederates say the wrong one to see if participants conformed or not

• findings : roughly one-third of participants conformed to the groups response, presence of a “true partner” decreased conformity

Pager (2003)

• research question: how does a criminal record impact callback rates for black and white job seekers?

• Had three hypotheses: criminal record will/will not significantly decrease callback rates for Black applicants/ will equally decrease callback rates for Black and White applicants

• Independent variable (s): criminal record and race

• Dependent variable: percent of applicants that received callbacks from employers

• Presented both white applicants (one with a criminal record and one without) in one group and did the same for the Black applicants

• Findings: having a criminal record negatively affected callback rates for Black and White applicants, even with a criminal record White applicants still received more callbacks than Black applicants with criminal records

survey weaknesses

• Less control over participants environment

• The salience and effect of a manipulation may be weaker 

• Often rely on self-report measure, which might not capture actual behaviors

field experiments

an experiment that takes place in the “real” world of a natural environment

What type of experiment was the Asch conformity experiements?

Lab experiment

What type of experiment was Pedulla and Thebaud’s Work Family Ideals?

Survey experiment

• Work family ideals

• Pedulla and Thebaud (2015)

• Used a between subjects design with three conditions (high constraint, medium constraint, low constraint)

• Used relationship options on survey as manipulation, showing different ones based on condition

• Found that men and women preferred egalitarian structure when available, presence of supportive policies sig. increased % of women who chose egalitarian but not men

audit atudies

a type of field study that is used to asses levels of discrimination in hiring, housing, admissions, etc

field experiments strengths

• Increased confidence in the generalizability of the findings 

• Good at testing more applied questions or testing success of interventions, not as much for testing abstract theories 

• Potentially lower risk of demand effects, especially when using double-blind design

field experiments weaknesses

• Less control over environment = higher risk of confounding variable impacting results

• It may be challenging to ensure to consistent exposure to the treatment

• Samples are not always random

• Can be challenging to replicate

• Additional ethical considerations to account for (e.g. access to medicine)

attribution theory

explains how people interpret the causes of behavior and events, distinguishing between internal (dispositional) and external (situational) factors

genetic essentialism

is the belief that genetic factors are the primary determinants of human behavior, identity, and social characteristics, often leading to misconceptions about the nature of these traits

Direct treatments

The manipulation or intervention is what it appears to be the participant, participant is likely aware of what they are being “exposed” to

Indirect treatments

The manipulation or intervention is not clear or apparent to participants. Purpose is to ‘prime’ something through a task that seems unrelated to the DV

Phelan (2003)

• research question: if a stigmatized character is attributed to genetic factors (as would occur increasingly with increased geneticization), how are peoples reactions affected?

• Hypotheses: 1. Attribution Theory: geneticization of a condition will results in lower levels of stigma. Stigma operationalized through blame, anger, punishment (negative), sympathy, helping (positive). 2. Genetic Essentialism: geneticization of a condition will result in higher levels of stigma. Stigma operationalized through beliefs, perceived differences, persistence, dangerousness, risk to family members; behavioral consequences:social distance, reproductive restriction.

• Phelan claims it’s possible for both these theories to be correct