is a statistic that helps you understand the strength of the linear relationship between two variables.

essence all propositions are subjected to an empirical (scientific) test   

Empiricism

practice based on experiment and observation; doctrine that all knowledge is derived from sense-experience

Goals of Behavioral Science (DEPF):

 description, prediction, finding causes, explanation

Basic Research

 fundamental questions about nature of behavior, often tests theories of behavior

Applied Research

issues in which there are practical problems/potential solutions; often guided by basic theory and research

Hypothesis

 statement that asserts about what is true in particular situation, often a statement asserting that two or more variables are related to each other

Prediction

 statement that asserts concerning what will occur in particular research investigation – ex/ schizophrenia is caused by a virus

 set of explanatory statements about behavior that can be tested through empirical research organize/explain existing data, leads to new insight, supported when data consistent with theory, revised when new data are collected that reveal weaknesses, replaced by newer theories that provide better explanations – ex/ Darwin's theory of evolution

 provides an overview of the whole article

general overview of literature with specific research questions/expectations

Includes:  participants, research design, procedure

exempt from IRB review, studying normal educational practices, use of educational tests (ex/ aptitude/achievement measuring), survey and observation of public behavior except when subject might be identified/sensitive behavior studied, archival research using existing data

standard psych measures and voice recordings, studies of cognition/perception, full informed consent is not generally necessary, but debriefing/other ethical concerns important

 research that may involve physical/psychological stress, invasion of privacy, measure of sensitive info where subject may be identified requires full review by IRB

 cost=potential risk to participants, benefit=participants, science, society 

participants agree to participate after being informed about purpose of study, potential risks/benefits, rights to refuse/terminate participation importance increases with increasing risk to participant

after study, inform participants about research questions being addressed, and why deception, if used, was needed

any event, situation, behavior, or individual characteristic that varies, has at least two values ex/ gender: male or female

definition of variable in terms of operations/techniques the researcher uses to measure/manipulate it

 behavior is observed as it naturally occurs, variables are only measured  problems: direction of cause and effect (ex/ does x cause y?), uncontrolled third variable (ex/ maybe z causes x and y?) 

 first IV then DV measured (cause effect) x causes y, not y causes x

 were variables adequately measured/manipulated? look at operational definition of variables that were studied

 can I infer that x caused y?  Higher in experimental method

can I generalize beyond the study? Other types of participants and ways of studying variables

 obtained score reflects true score and measurement error – ex/ 0.04 seconds = true score and any measurement error reliable measures have little measurement error

 a single score reflects true score but ex/ a bit faster today because good night's sleep, a bit slower cause you missed lunch

 multiple measures to average measurement error, not possible for most research  use measures that have little measurement error = reliability 

 take measure twice, reliability=when two scores are similar

estimation we use our single measurement instrument administered to a group of people on one occasion to estimate reliability. In effect we judge the reliability of the instrument by estimating how well the items that reflect the same construct yield similar results. We are looking at how consistent the results are for different items for the same construct within the measure. There are a wide variety of internal consistency measures that can be used.

 sometimes rater/judge can make an observation of behavior examines agreement of observations made by two or more raters/judges

 combine multiple measures, careful recording procedures, clear and unambiguous questions

degree to which measurement device appears to accurately measure a variable 

... do scores on measure relate to other measures in expected ways? – Ex/ do people with high self efficacy predict that they will perform better on task?

... is the actual general agreement among ratings, gathered independently of one another, where measures should be theoretically related.

degree to which a measurement device accurately depicts behavior on criterion measure – ex/ paper and pencil measure of leadership – ability predicts actual leadership behavior 

 a measure is reactive if it changes the behavior being measured, useful to use unobtrusive measures to complement studies using more reactive measures

variables that have two or more categories, no numeric scales – ex/ male/female

rank order points on scale, numeric values limited – ex/ 2, 3, or 4-start restaurants  intervals between items are unknown and are not equa

numeric properties are important, values such as 1, 2, 3… used in calculations. ex/ score on personality test & assume equal amounts between numbers & no true zero amount

 has true zero point (absence of variable) use more sophisticated statistical tests for ration and interval scales 

 the subject knows they are being observed and measured but they do not know for what 

Employs questionnaires and interviews to ask people to provide information about themselves - their attitudes, beliefs,  demographics, etc...

 A pattern of individual response to questions on a self-report measure that is not related to the content of the question.

The defined group of individuals from which a sample is drawn

You can specify the probability that any member of the population will be included in the sample

Used when you can’t identify individuals

Randomly sample clusters of people in identifiable groups, e.g., classes in a school

Used when the difference between groups is not relevant 

A type of sampling procedure in which one cannot specify the probability that any member of the population will be included in the sample

Percentage of those sampled that complete the survey

Study the sample people at two or more points in time

 the participants that are in the study that know what is going on and manipulate the real participant.

 really good valid, external and internal good validities.

Posttest-only design

Pretest-Posttest design

Random assignment to conditions

Experimental control of all other variables

 we manipulate independent variable and test the dependent variable.

 when they come in they are tested. (within subject or repeated measures design)

 compare the control group from the experimental group . 

 testing the same group or person 

 same individuals in all conditions 

1: prevent researchers from subconsciously letting their opinions or preferences affect their decisions about what participants gets exposed to what

2: even distribution on that normal curve on our individual participants. SO we won’t end up with same characteristics we want an equal chance.

3: lets us derive probabilities . It is the probability that if i go look at the people in the real world they would act the same way that responded in the experiment. 

three reasons why we randomize our designs:

The process of choosing members of a population of interest

The defined group of individuals that the researcher is interested in.

When we have a ___________: survey, systematic observation, archival research 

When we have a _____________: experimental design (usually with a manipulation)

analyzing an existing data.

The researcher manipulates the independent variable, to create groups, that differ in the levels of the dependent variable.

 allows a known probability that each potential participant will be chosen. 

Divide population into strata

A type of sampling procedure in which one cannot specify the probability that any member of the population will be included in the sample

Whoever you can get to take part in your study.  You don’t use any special techniques to find your subjects.

is obtained according to the discretion of someone who is familiar with the relevant characteristics of the population. 

Naturalistic observation- also called “field observation”

In this method, the researcher collect data over long period of time using variety of techniques. 

provides a description of an individual (person, business, neighborhood…) that is of some kind of importance.

When we need to find out information without asking the subjects directly or when the subjects are not people (like movies, texts etc.), we must use  (used with systematic observation or archival research)

Focuses on specific behaviors in a narrow context, which the researcher is interested in

already developed prior hypotheses about the behaviors.

Existing data sources such as:

Statistical records

Survey 

Written and mass communications

Researchers conducted an in-depth study with certain 9/11 victims to understand the psychological impact of the attack on the World Trade Center in 2001

Researchers recorded the time it took drivers in college parking lots, to back out of a parking stall and compared the time when another car was waiting to the time when another car was not waiting for the space

Researchers examined unemployment rates and the incidence of domestic violence police calls in six cities

Contents of mate-wanted personal ads in three major cities were coded, to determine whether men and women differ in terms of their self-description

The researcher spent over a year meeting with and interviewing Aileen Wuomos, the infamous female serial killer who was the subject of the film Monster, to construct a psychobiography

a variable that varies along with the independent variable, so it is not possible to determine which of the variables is responsible for the observed effect!

After reading about a less severe crime the more severe crime might seem much worse

1. practice 

2. fatigue

 all possible order of conditions are included in the experiment

a limited set of orders constructed to ensure that: 1. each condition appears in each ordinal position 2. each condition preceded and follow each condition at least one time (number of orders is equal to number of positions)

the subjects are matched according to important variable and then the couples are divided equally among conditions. 

Assign to all levels of the independent variable

Randomly assign to only one level of the independent variable

count on occurrences of phenomenon (ex/ examining pictures in children's books- boys/girls and what categories of activities they are engaging in)

distance between the true population value and the mid-point (mean) of the sampling unit indicates the amount of bias in the sample 

 two independent variables with two levels of each variable. One dependent variable

What are the outcomes in a 2 by 2?

 tests whether a categorical independent variable has an effect on some continuous dependent variable

that the independent variable and the dependent variable are mutually related 

Temporal precedence

Did X come before Y?  Did our independent variable occur before our dependent variable? 

this is when you non-randomly assign participants to an experimental condition and a control condition.  It is a between-subjects kind of study.

Interrupted Time Series Design

multiple pre- and post measures are taken and compared.  However, a control group is also tracked, allowing for a “between groups” comparison

similar to plain reversal designs, except that the treatment is introduced at different times for different subjects or targeting different behaviors for a single subject

Comparison of multiple periods, designated by letters (A, B, C) in which treatment is alternatively either absent or present.  

The design attempts to establish the link between the administration of the treatment and a specific behavior change

Study people of different ages at one time

Problem: cohort effects may explain age differences

Study the same people as they grow older

Examine changes with age

Expensive and difficult

Combine cross-sectional and longitudinal approaches

Take cross-sectional approach at time 1 but follow up the same people for a longitudinal study at time 2

Provides immediate age difference information

 simplest indicator, the most frequent score

the middle score in a distribution, very useful if scores are skewed in one direction

 arithmetic average dividing the sum by the number of scores

Refers to relationship between two variables

A statistic that describes how strongly variables are related

 provide a way of making a decision about the population means

A statistical significance test to evaluate the null hypothesis of no difference in the population means of two groups

probability of failing to reject the null hypotehsis that is actually false

Incorrectly accepting the null hypothesis when there really is a difference

Probability of correctly rejecting the null hypothesis

1-Beta

young

high in cognitive skill

high in need for peer approval

and are easy to obtain

Rats are hardy, cheap, easy to rear, and well adapted to laboratory existence

Easy to obtain

social

highly educated

in need of approval

Volunteers differ from nonvolunteers in that volunteers tend to be more:

Half of the participants are given the pretest; the other half receive the posttest only

The extent to which the experiment has an impact on the participants, involves them, and makes them take the experiment seriously.

The extent to which the independent variable manipulation is similar to events that occur in the real world.

Repeating a research study to determine whether the results can be duplicated

Replication of research using the same procedures for manipulating and measuring the variables that were used in the original study

Replication of research using different  procedures for manipulating and measuring the variables that were used in the original study

A set of statistical procedures for combining the results of a number of studies in order to provide general assessment of the relationships between variables

researchers can overestimate/underestimate how often something happens – tend to confuse what they believe they see and what they want to see with what actually happened 

specify where the population value probably lies

the spread (variability) of your sampling responses  

Non-Equivalent Control Group Design

 participants are not randomly assigned -> usually self-selected groups -> problems of selection differences 

task if so hard that almost no one can perform well

 participants quickly reach maximum performance level since task was too easy

 does the effect of the IV depend on the particular level of the other IV? – Ex/ distraction while standing is bad for introverts, however, distraction does not bother extravert

 a variable moderates the effect of an IV

 may occur whenever the experimenter knows which condition the participants are in 

Presenting written, verbal or visual material

Instructions or stimulus presentation

most common with social psychology 

Use of confederates, usually appears to be another participant in the experiment.

attitudes, liking, intended behavior, emotions….

 recording of responses of the body

 direct observations of behaviors

 make participants think that the experiment is studying one thing when actually it is studying something else (for example – Asch’s study)

 a “trial run” with a small number of participants

 “Where are you originally from?”

North America

South America

Europe

Africa

Other

When the topic being asked is too personal or sensitive, we try to ask the questions in a projective way 

Rare, expensive adminstration

Most common for large surveys 

Tend to be qualitative and open-ended

Explain exactly how the study was conducted. 

Maintain the temporal sequence of events so the reader is able to visualize what the participants experienced.

This section tells the reader what instructions was (or wasn’t) given to the participants. 

Include any detail that might be important in a replication of the study 

Common sense

Observation of things happening in the world

Past research

Practical problems

Theories

The research process->

sources of ideas:

Research question

Hypothesis

Prediction

Hypothesis testing, results and discussion

Four steps of scientific research in social sciences:

Introduction - Literature review

Introduction – Hypotheses

Prediction

Research method

Results

Discussion and conclusions

Reference list

Sections in a journal article:

A statement about 2 or more variables

A clear phrasing that can be understandable in one way only

Can have only 2 outcomes: true or false

Must be studied empirically

4 conditions for hypothesis

Time organization: what happened before what?

Dismissing alternative explanations

There is a strong statistical connection between the variables

3 conditions for causality

The problem under study- 2-3 sentences: broad context and significance of the topic being study

The literature review- a description of past research and theory. The basis of what we’ll investigate.

The rational (connects between the literature review and the hypothesis), hypothesis and prediction of the study 

Just getting a treatment may be sufficient to cause an observed change in behavior/feeling

Consistency of behavior

Run all conditions simultaneously

Automated producers (computers)

Experimenters who are unaware of the hypothesis

The method of tenacity –or the ostrich method

Clinging to beliefs that have been around for a long while and excluding anything that might alter or challenge those beliefs

The method of authority

 believing something because someone with “authority” tells you it is so.

relying on individual powers of reasoning and logic to make sense of the world and to explain it to others