PY 355 Final Exam
Terms in this set (96)
Do not perform any mathematical operation.
a. Ex: Males = 1, Females = 2
The rank ordering people's behaviors or characteristics.
a. Ex: The other in which runners complete a race.
Makes inferences about populations using data drawn from the population (sample).
Summarize a given data set, which can be either a representation of the entire population or a sample of it.
A procedure used to examine every study that has been conducted on a particular topic to assess the relationship between whatever variables are the focus of the analysis.
The portion of the total variance in participants scores that
Statistic used to indicate the amount of variability in participants responses.
The portion of the total variability in participants' scores that is related in an orderly, predictable fashion to the variables the researcher in investigating.
A measure of the strength of the relationship between two variables; indicates the proportion of the total variance that is systematic variance.
Total Sum of Squares
The sum of the squared deviations of the scores from the mean.
How much each score differs from the mean
Sum of a set of scores divided by the number of scores; asses variability by seeing how much the scores vary around mean
e.g., scores tightly clustered -> small variance, scores spread out -> large variance.
The difference between the highest and the lowest scores.
Say you are trying to describe a distribution of scores, the shape and the level of variability. Why can the average of a set of scores be misleading? Why might the range be misleading?
Range and mean can be offset by outliers.
Equal differences between the numbers reflect equal differences between participants, but there is no true ZERO POINT.
E.g., Scores on an IQ test, ratings on a 5-point scale, time of day.
Contains a true zero point.
e.g., weight, the number of questions answered correctly, the time it takes to complete a task, annual income.
Expresses the strength of the relationship between two measures.
a. Can range from -1.00 to +1.00
b. Correlation of .00 indicates no relationship between the variables.
c. The sign indicates whether the relationship between the variables is positive or negative (inverse).
The extent to which a measure of a hypothetical construct relates as it should to other measures.
Scores on a measure are related as expected to a criterion that is assessed at the time the measure is administered.
e.g., An embarrassability scale (administered today) predicts blushing in the current situation.
The extent to which a measure appears to measure what it's supposed to measure.
Entities that can't be directly observed but are inferred on the basis of empirical evidence.
e.g., intelligence, impulsivity, status, motivation, love, self-esteem, attachment style.
The degree to which a measurement procedure actually measures what it is intended to measure rather than measuring something else (or nothing at all).
The consistency or dependability of a measuring technique. Reliability = true variance divided by total variance.
a. Test-Retest Reliability: Consistency of participants' responses on a measure over time.
b. Inter-Item-reliability: Assess the degree of consistency among the items on a scale.
c. Split-half Reliability: Divide the items on a scale into two sets and examine the correlation between the sets.
What factors must be considered when selecting the methods for an observation study (3 important decisions to make, and the characteristics associated with each)
1. Will the observation be in a naturalistic or contrived setting.
2. Will participants know they are being observed?
3. How will participants' be recorded
Compared Naturalistic Observation and Contrived Observation
Naturalistic: observe behavior as it occurs naturally with no intrusion or intervention by the researcher.
- Researches observe people or animals in their natural environment and record their behavior.
- Participant Observation - one type of naturalistic observation: researcher engages in same activities as the people he/she is observing.
Contrived: behavior is observed in situations set up specifically for being able to observe that behavior.
- Most contrived observations take place in the lab; however, some researches set up situations outside the lab to observe people's reactions.
What allowed Conger and Killen (1974) to predict the proportion of the time the participant would direct their attention towards one of the two group discussants?
Found time spent talking matched rate of agreement from the listener. Showed that time allocation to other people in dinner conversation matched the frequency of reinforcers provided by these people. The relative amount of time the volunteer looked at each confederate.
Measure of Latency
Another way of defining "reaction time." A measure of the time that elapses between the presentation of a stimulus and the participants response.
The time elapses between the presentation of a stimulus and the participant's response. (e.g., rat video - time it takes rat to get through trap while on different drugs)
Task completion time
The length of time it takes participants to solve a problem or complete a task.
The time that elapses between two behaviors.
How long a particular behavior lasts (e.g., how long people talk during a convo; how long people engage in eye contact)
Observational Rating Scale
Researcher rates the quality or intensity of a certain behavior (e.g., rating a chi's crying as (a) slight, (b) moderate, or (c) extreme.
What name is given to the measure of how consistent observational ratings are between or among multiple observers or coders? What steps can be taken to ensure reliability?
- Make sure each observer / coder is properly trained to ensure reliability.
- Observational Rating Scale: each researcher rates the quality or intensity of a behavior.
- Must have a clear and concise operational definitions for all behaviors that will be observed and recorded.
- Rates should practice using a coding system by comparing and discussing their coding ratings.
Sample / Sampling
The process by which a researcher selects participants for a study. There are several different ways to select samples from the population.
A sample that is selected such as the likelihood that any particiular individual in the population will be selected for the sample can be identified. Probability samples are rare in psychology because the purpose of most behavioral research is NOT to describe how a population behaves; rather, psychologist test hypotheses about how psychological variables relate to one another and such studies don't require probability samples.
Non probability sample
Researches do not know how the probability that a particular case will be chosen for the sample, it can't be determined.
Error of estimation can't be calculated. Most research involves this type of sampling. It's a valid method when the goal is to test hypotheses regarding how particular variables relate to behavior rather than describe a particular population.
Used if the goal is to make inferences about a population if the goal is to make inferences about a population, then a representative sample must be used.
The extent to which characteristics of individuals for the sample differ from those of the population. Because of sampling error, results obtained from sample differ from what would be obtained using the whole population.
Error of Estimation
Indicates the degree to which the data obtained from the sample are expected to deviate from the population.
- Its important only if the sample is a probability sample.
- Function of three things: sample size, population size, and variance of the data.
A list of the population from which the sample is to be drawn.
Stratified Random Sample
The population is divided into strata, then the participants are randomly selected from each stratum. Ensures that researchers have an adequate number of participants of each stratum.
A subset of the population that shares a particular characteristics (such as race, gender, location).
Proportionate Sample Method
A variation of stratified random sampling in which cases are selected from each stratum in proportion to their prevalence in the population.
Use whatever participants are readily available. Researchers often use convenience samples of college students. Because college sample may differ from the population at large, some results may not generalize to all people.
Researches use their judgment to decide which participants to include in the sample, trying to choose respondents who are typical of the population.
Convenience sample in which the researcher takes steps to ensure that certain kinds of participants are obtained in particular proportions.
e.g., researcher may wish to obtain an equal amount of men and women.
The ability of a research design to detect any effects of the variables being studied that exist in the data.
A sample that provides reasonably accurate estimate of the population at reasonable effort and cost. Researchers usually opt for this.
Sample groupings or clusters of participants. Clusters are based on naturally occurring groups that are usually close in proximity.
Simple Random Sample
Most common probability sample; every possible sample of the desired size has the same chance of being selected from the population.
Involve taking every so many individuals for the sample.
Cross-Sectional Survey Designs
Sample consists of a "cross-section" of the population at a specific point in time.
Successive Independent Samples Survey Design
;Two or more samples of respondents answer the same question at different points in time. Looking for participants reliability.
Longitudinal Survey Design (Panel)
A single sample of respondents is questioned more than once; dropouts may be a problem.
Used when the variable on the x-axis is on an interval or ration scale of measurement (y-axis = frequency of each score)
(Line Graph) Axes are labeled as they are for the histogram but lines are drawn to connect the frequencies of the class intervals.
Used when the variable is on a nominal or ordinal scale of measurement.
Shows how much one variable is affected by another. The relationship between two variables is called their correlation.
Bars above and below the means in a graph indicating researcher's confidence in the value of the mean.
Measure of Central Tendency
Describes the way in which a group of data cluster around a central value; a way to describe the center of a data set. There are three measures of central tendency: the mean, the median, and the mode.
Mathematical Average; most commonly used measure.
The middle score of the distribution.
Most frequent score.
Represents the distribution of many random variables as asymmetrical bell-shaped graph.
Confidence Interval (CI)
Measures the probability that a population parameter will fall between two set values; the most common being 95% and 99%.
More low scores than high scores.
More high scores than low scores.
Standard Deviation (SD)
The square root of the variance, generally easier to interpret.
Describes a particular participant's score relative to the rest of the data. Indicates how far the participant's score falls from the mean in terms of SD.
What percentage of a normal distribution falls within 1 SD (i.g., above and below) of the mean? What percentage fall outside + 3 SD?
- 68% falls within 1 SD above and below the mean
- 99.7% falls within 3 SD above and below the mean
How do the correlation coefficients reflect the strength of a relationship between two variables?
- Can range from -1.00 to +1.00
- The sign indicates whether the relationship between the variables is positive or negative.
- Positive correlation - A direct positive relationship between two variables; as one variable increases, the other one increases as well.
- Negative Correlation - An inverse negative relationship between two variables; as one variables increases, the other decreases.
On a scatterplot, how might a positive, negative, or zero correlation be reflected by the pattern of the data?
When there is a perfect correlation all the of the data will fall in a straight line.
- Positive: upwards slop to the right
- Negative: Slope downwards to the right
- Zero correlation appears as a random combination of dots.
If the correlation between two variables is .00, what might you conclude? Could there be a non-linear relationship between the variables?
A correlation of .00 indicates there is no linear relationship between the two variables.
A partial correlation is different than a correlation in what way?
Partial Correlation is the correlation between two variables with the influence of one ore more other variables statistically removed.
Correlational Research is used to describe the relationship between two or more naturally occurring variables.
How would you describe a directional vs. non-directional hypothesis?
Directional: predicts the direction of the correlation (positive or negative)
Nondirectional: predicts that the variables will be correlated but does not specify whether the correlation will be positive or negative.
Regression analysis can be used to do what?
In the equation y = mx + b what is the predictor variable, the regression constant, and the regression coefficient?
Predicator Variable - x; is the variable we used to predict y
Regression Constant - B; y-intercept of the line that best fits the data
Regression Coefficient - m; slope of the line that best represents the relationship between x and y.
Rationale for using a factor analysis
Used to identify the factors that account for the relationship between variables in a set.
How many groups must there be in an experimental design? What about a design in which there are two levels of an independent variable and the researcher wishes to include a control group?
There must be at least two groups of subjects in an experimental design.
- Experimental group: participants in an experiment who receive a nonzero level of the independent variable.
- Control Group: Participants in an experiment who receive a zero level of the IV (or the absence of the variable of interest)
What is the distinction between a subject variable and an independent variable? What is the distinction between experimental design and a quasi-experimental design?
- Participant (subject) variable: a personal characteristic of research participants such as age, gender, self esteem, weight , or extraversion.
- Independent: a variable (often denoted by x) whose variation does not depend on that of another.
- Quasis-Experimental Design: researcher cannot assign participants to conditions an/or manipular the IV. Instead, comparisons are made in a group that already exists before and after the treatment.
Experimental: a controlled experimental factor is subjected to a special treatment for purposes of comparison with a factor kept constant.
How might an experimenter control for order effects in a repeated-measures design?
Order effects occur when the effects of a particular experimental condition are contaminated by its order in the sequence of experimental conditions in which participants are tested.
- To protect against order effects, researchers use counterbalancing. Counterbalancing involves presenting the levels of the IV in different orders to different participants.
- Practice effects - participants' responses are affected by completing the dependent variable many times
- Fatigue Effects - participants become tired or bored as the experiment progresses
- Sensitization - participants gradually become suspicious of the hypothesis as the experiment progresses
In Fournier 2010, what did they have to say about the effectiveness of placebo vs. antidepressant medication?
- No significant difference in placebo vs. antid
- Failed to account for baseline level of depression (only compared final results)
researchers use special terms to describe the size and structure of factorial designs
- A 2x2 is a factorial design with two IVs each with two levels.
- A 3x3 factorial has two IVs each with 3 levels
- A 2x2x4 has three IVs, two with two levels, and one with four.
Randomized Group Factorial Design
Participants are assigned randomly to one of the possible combinations of the IV.
Matched Group Factorial Design
Participants are first matched into blocks on the basis of some variable that correlates with the DV. The participants in each block are then randomly assigned to one of the experimental conditions.
Repeated Measures Factorial Design
Each participant participates in every experimental condition.
Mixed Factorial Design
Participants are randomly assigned only to one level of some IV but receive every level of other IV, also called between within design.
Know how to distinguish between main effects and interactions when the results are verbally described as well as visually when results are graphically displayed.
- The main effect of an IV is the effect of that IV
- Interactions on a graph appear as non-parallel lines.
What is a mixed design? (aka expricorr design)
Mixed factorial designs are designs that include both indepdent variables (that are maniuplated) and participant variables (that are measured).
- The mixed design include both the independent variables and the dependent variables. It is used to determined whether effects of the independent variable generalize only to participants with particular characteristics, examine how personal characteristics relate to behavior under different experiemntal conditions, and reduce error variance by account for individual differences among participants.
Inferential Statistics are different than descriptive statistics in what way?
Inferential: mathematical analyses that allow researchers to draw conclusion regarding the reliability and generalizability of their data (t-tests)
Descriptive: numbers that summarize and describe the behavior of the participants in a study (mean, sd)
The hypothesis that the independent variable will not have an effect on the dependent variable.
The probability of making a type 1 error (and erroneously believing that an effect was obtained when it was actually due to error variance)
The probability of making a Type II error (and erroneously failing to find an effect that was actually present)
Type 1 Error
A researcher rejects the null hypothesis when it is true.
Type 2 Error
A researcher fails to reject the null hypotheses when it is false
Null hypothesis is false and you have rejected the null hypothesis.