Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
124 Cards in this Set
- Front
- Back
|
Alpha |
-This is the confidence level we use in expressing significance. -alpha= .05 -We express significance in our interpretations as p<0.05 |
|
|
Alternative Hypothesis |
-The hypothesis we hope to support in our research; -predicts a significant relation or difference exists between the groups we are comparing. -Opposite of the Null hypothesis |
|
|
One Sample t-Test |
-1 group -Measures Mean and Standard Deviation |
|
|
Independent Samples t-Test |
-2 Groups - Measures IV+DV "t for 2" |
|
|
One Way ANOVA |
-3 Groups -Measures Many scores/group -1 IV -Independent Groups -F values |
|
|
Repeated Measures ANOVA |
-3 Groups -Measures Many Scores/Group -1 IV -Repeated Measures - F Values |
|
|
ANOVA- ANalysis Of VAriance |
An inferential parametric statistic we use for comparing the means of a few groups (F for a few-- 3 or more). We symbolize it with the letter F- developed by Sir Ronald Fisher, the F references his last name. |
|
|
APA Style |
The writing style that has become the standard for professional writing in psychology, in most of the social sciences, in many of the natural sciences, and in some of the humanities. It is detailed in a Publication Manual and deals with both broad qualities of writing and with specific details. |
|
|
Average Deviation |
-A measure of variation or spread - indicates average difference between scores and mean in distribution -not a good theoretical measure of variation,; standard deviation is more useful |
|
|
Bar Histogram |
A graph of a frequency distribution in which the vertical bars represent the frequency of each score or group of scores in the distribution |
|
|
Bell Curve |
Guassian Curve or standard normal frequency distibution |
|
|
Between Groups ANOVA |
An inferential parametric statistic -compares the means of three or more groups -each group is composed of different participants -AKA: Independent Groups ANOVA or Simple ANOVA |
|
|
Between Groups |
-Comparing differences between groups produced by IV against the differences due to chance *F= MSbetween/MSwithin *F=MStreatment/MSerror |
|
|
Box Plot |
A graph that depicts the overall distribution of a data set. -Lower End= 1st Quartile -Upper End= Third Quartile -Middle line= Median - Whiskers= Minimum/Maximum Score *Simple way to display the center of a set of scores plus the spread of the scores |
|
|
Central Limit Theorem |
Normal distribution is ubiquitous. It shows up everywhere when we measure peoples' behavior, no matter what we measure. |
|
|
Central Trend |
A descriptive statistic that represents the center of a data set; -the value that all the data seem to gather around -The mean, median, and the mode are commonly used as measure of this |
|
|
Citation |
The statistical information that appears at the end of any conclusion or interpretation we write. Example: t(14)=2.78, p<.05 |
|
|
Cohen's Conventions |
-Guidelines for all effect sizes - general regions and the language used for describing the effect, strength, or variance explained .0= negligible, paltry, very little or none (99% overlap) .2= weak, small, not much, little bit (80% overlap) .5= moderate, modest, some, a fair amount (50% overlap) .8= strong, lots, a large amount, (20% overlap) |
|
|
Cohen's d |
-Measure of effect size -Measure of how much difference exists between two groups measured by a t-Test -Measures how much effect the independent variable has on the dependent variable - d-value gives suggestions for the language you use for expressing how much of a difference or size of effect |
|
|
Confidence Level |
We are never CERTAIN a conclusion is right, so we express our confidence in the conclusion. p<.05 Probability our conclusion is wrong is less than 5% |
|
|
Confidence Interval (descriptive) |
The range within which the true population is likely to lie. -Interval that surrounds the sample mean -cited in brackets |
|
|
Confirmation Bias |
Our unintentional tendency to pay attention to evidence that confirms what we already believe and to ignore evidence that would disconfirm our beliefs. -we use descriptive statistics to protect against bias |
|
|
Confounding Variable |
-A variable that systematically covaries with the IV so that we cannot logically determine which variable is at work on the DV -Problem for interpreting research -Methods to eliminate them include: balancing; counterbalancing; and polishing |
|
|
Convenience Sample |
A sample of readily available people, not a random sample from the entire population of interest |
|
|
Continuous Variable |
A variable that falls along a continuum and often has fractional amounts (decimal numbers). -Unlimited number of possible values -Time, winnings, weight |
|
|
Control Group |
A level of the IV that matches the experimental group in every way but the experimental manipulation. -The baseline against which we compare the the experimental group |
|
|
Counterbalance |
-minimize order effects by varying the order presentation of different levels of the IV |
|
|
Critical Value |
-the value of the test statistic that is the dividing line between normal and deviant and a sampling distribution -We use it to determine significance if the statistic we obtained from our data is beyond the CV, we say the outcome is significant |
|
|
Degrees of Freedom (df) |
-the number of scores in a sample that are free to vary - n-1 - or n-2 - or something less than the number of groups |
|
|
dependent variable (DV) |
-Usually our data |
|
|
Descriptive Statistics |
-statistics that describe a distribution by providing information about central trend, its width and its shape -mean, median, mode, range, SD, Skew |
|
|
Discreet Variables |
-Variable measured in whole numbers -limited number of values -gender, # of speeding tickets, and personality tests |
|
|
Dot Plot/Dot & Whisker Plot |
-a graph where a dot is used to represent the mean of a set of scores -a whisker to represent one SD above and below the mean |
|
|
eta squared |
-a measure of effect size, like Cohen's d - it tells how much variability the DV is explained or accounted for by the IV. |
|
|
Effect Size |
-the proportion of variance in the DV that is explained by the IV |
|
|
Error |
-a mistake -everything unaccounted for in the research |
|
|
Error Bar Graph |
-a graph of cell means that is often used when there are three or more groups in the analysis -the dot= mean -line above/below= 95% CI for the mean -Useful because it allows us to see overlap and compare similarity |
|
|
Estimate |
-Making estimates about the population based on our experience with samples 1. We commonly make (1) point estimates which is the mean cited in parenthesis. 2. Interval Estimates- which is the 95% CI cited in brackets 3. Predictive estimates of why in regression analysis |
|
|
eta-squared |
-an inferential statistic for measuring effect size with ANOVA |
|
|
Experimental Group |
- a level of the IV that receives the treatment or intervention of interest in the study -We often compare it to the control group |
|
|
Experimental Method |
-a research method that allows us to find a cause and effect -We can manipulate variables, control the situation and use random assignment - the manipulated variable is the IV and we measure its effect on the DV -the IV is sometimes called a situation variable because it is that feature of the situation that we manipulate This is usually referred to as a TRUE experiment to distinguish it from a Quasi experiment |
|
|
Exploratory Data Analysis (EDA) |
-an effort to summarize data, often with visual displays -sometimes described as data mining because rather than testing a specific hypothesis (the null) we just dig around in the data to see what it looks like. -And to see if there are any Hypotheses that might be interesting to test |
|
|
Extraneous Variable |
-a randomly distributed influence that detracts from our efforts to measure what we intend to measure -we usually refer to the effects of this as error because we can't possible eliminate all extraneous variables -we compensate by taking error into account in a statistical measures |
|
|
Factor |
-A term used for an IV in a study with more than one IV |
|
|
Figure |
-Found in APA writing -any visual presentation of data, like a photograph, drawing, or graph |
|
|
First Quartile |
-The 25th percentile of a data set -bottom of box on box plot *the entire box goes from 1st to 3rd Quartile |
|
|
F-Ratio F= MSbetween/MSwithin F=MStreatment/MSerror |
-the formula for the F test is the ratio of between-group variance to within-group variance. -also expressed as the ratio of treatment variance to error variance -the top number in the ratio reflects difference do the IV + random, chance differences. The bottom number in the ratio reflects random, chance differences *If IV has no effect on DV, F=1.00 *IF IV has an effect on DV, F> 1.00 |
|
|
Frequency Polygon |
-A smoothed line graph of the frequencies of each score. -y axis= frequencies -x axis= individual scores -most common= bell curve or standard normal frequency polygon |
|
|
Generalizability |
-our ability to apply findings from a study of one sample or in one context to other samples or contexts. -our ability to apply findings from one sample to the population the sample is assumed to represent |
|
|
Grand Mean |
-Mean of means -the mean of the scores for all participants in a study |
|
|
Graph |
-a pictorial display of quantitative information -a figure |
|
|
Heterogeneous |
-Consisting of dissimilar elements -When a set of scores has a large standard deviation, there is lots of variation from the mean and the scores are said to be heterogeneous |
|
|
Homogeneous |
Uniform or similar in kind. When a set of scores has a small standard deviation, there is little variation from the the mean |
|
|
Hypothesis Testing |
Testing whether a hypothesis is supported by the results of our research -involves setting criteria for judging significance, deciding on a one-tailed or two-tailed test, conducting the appropriate statistical analysis, checking a statistical table of critical values to determine the significance of the statistical analysis, and then either retaining or rejecting the null hypothesis. |
|
|
Independent Groups |
-Research with different people in each group of study -we could randomly assign any person to any one of the groups |
|
|
Independent Variable |
The variable in a study that we manipulate or select for. This variable usually identifies each group in the study and appears on the horizontal axis of a graph of the group means. |
|
|
Inferential Statistics |
Inferences are conclusions from data. -We use these statistics to draw conclusions about a population based on data we have collected from a sample. |
|
|
Interpretation |
A statement about the outcome of a statistical analysis. The statement is a conclusion about the outcome or an interpretation of the meaning of statistics. The interpretation usually addresses the research hypothesis or if the null hypothesis should be retained. *It is always concluded with a citation of the relevant statistical information |
|
|
Interval Estimate |
-A statistical term for setting an interval around the sample mean which we believe will contain the true population mean. *think you speed on the highway compared to the actual speed limit- how fast can I go without getting a ticket? |
|
|
Interval Scale |
-a scale of measurement in which the units are all equal in size, but the zero point is arbitrary. -body temperature, GSR on Lie detector test, and some standardized test scores such as IQ |
|
|
Level |
-A discreet value or condition that an IV can take on -Seen in ANOVA when we have to put data into the computer organized by level names |
|
|
LSD Test |
-an inferential statistic for making pair-wise comparisons between two means in an analysis of variance. Pay attention to asteriks (*) in the Table, They mean significance |
|
|
Mean Square (MS) |
-An estimate of variance: -between groups; within groups; or total variance -MS is not square of the mean; it stands for the mean of the squared deviations, which is variance. |
|
|
Measures of Central Trend |
A number that describes a set of scores by indicating its center or middle. -mean, median, mode |
|
|
Measure of Variation |
A number that describes how much scores are spread around the mean of a distribution. Common measures of variation are the standard deviation, the variance, and the range |
|
|
Median |
-a measure of central trend -the middle score in a set of scores after the scores have been arranged from lowest to highest. -Not affected by extremes - |
|
|
Mode |
-a measure of central trend; - the score in a distribution that is the most frequent -Occasionally more than one score occurs with highest frequency and we talk about the distribution as bimodal, or perhaps even multimodal. -There might be two or more underlying distributions |
|
|
Negative Skew |
A distribution in which the peak is to the right of the center point, so the scores are piling up at the top of the distribution, and the tail extends toward the left, or in the negative direction. We express skew as negative or positive not left or right. |
|
|
Nominal Scale |
A scale in which objects or individuals are divided into categories and numerals are assigned to the categories just to name them. -apply simple count |
|
|
Normal Curve |
A symmetrical, bell-shaped frequency polygon representing the normal distribution |
|
|
Null Hypothesis |
The hypothesis that says no difference or no relationship exists between the groups being compared. |
|
|
One-Way ANOVA |
-An inferential statistic test for comparing the means of a few groups (3 or more), usually using an independent groups or between subjects design. |
|
|
One-Way Repeated Measures ANOVA |
-a statistical test for comparing means of a few (3 or more) groups using a repeated measures, or correlated-groups, or within-subjects design. -Each participant is tested on all levels of the IV |
|
|
Operational Definition |
Defining a variable in terms of the operations (activities) we use to measure or manipulate it. -Benefit: know what we are talking about by how we measure it |
|
|
Order Effect |
-effect produced when a participant's behavior changes when the DV is measured repeatedly.
|
|
|
Ordinal Scale |
A scale on which individuals or attributes are categorized along a continuum of ranks. Common ranks we use are 1st, 2nd, 3rd... and Likert scales |
|
|
Outlier |
An extreme score that is unusually high or low in comparison with the rest of the scores in a sample. |
|
|
p Level |
The probability we use to decide on the critical values, or cutoffs, in hypothesis testing. We will always use p<.05 as our criteria for significance in this class, but psychologists in different areas of research and for different applications may us p<.01 or p<.001 |
|
|
Parameter |
A statistic based on the whole population; it is usually symbolized by a Greek letter. Common parameters are mu for the population mean and sigma for the population standard deviation. |
|
|
Parametric Test |
A statistic test based on assumptions about population parameters (or characteristics). The most common assumption we make inpsychlogy is that our measurement (the DV) is normally distributed and parametric procedures, such as CI, t, F, or r are warranted. |
|
|
Percentile |
The percentage of people of people who scored at or below a given raw score.
-Most often used in psychology to report a person's performance on a psychological test. |
|
|
Placebo Effect |
The result achieved when just the expectation of an outcome either causes or appears to cause that outcome to take place. |
|
|
Population |
-All of the people, representing by a sample in a study, to whom we mean to generalize. -This is not likely all of the people in the world |
|
|
Positive Skew |
-a distribution in which the peak is to the left of the center point, so the scores are piling up at the bottom of the distribution and the tail extends toward the right or in the positive directions. |
|
|
Post Hoc Test |
Post Hoc means "after the event." When we have a significant ANOVA, we can follow it by testing all possible pairs of groups to pinpoint which ones differ significantly form each other and which ones do not. *LSD |
|
|
P-rep |
A measure of the likelihood we can replicate the results of a study, ideally in a different context or with a sample that has different characteristics. Using our results from an analysis of a study we use the p value and put it into the excel formula |
|
|
Psychometrics |
The branch of statistics used in the development of psychological tests and measures. |
|
|
Quasi-Experimental Method |
A research method in which the IV cannot be manipulated. -Most often the variable is a characteristic of people such as gender that cannot be randomly assigned -Restricts the conclusions we can draw to difference conclusions |
|
|
Random Assignment |
- Conducting an experiment and insuring every participant in it has an equal chance of being assigned to any of the groups or experimental conditions |
|
|
Random Sample |
Taking a sample in such a way that every member of the population has an equal chance of being selected into the study |
|
|
Range |
a measure of variation calculated by subtracting the lowest score from the highest score in a distribution |
|
|
Ratio Scale |
The most sophisticated scale of measurement we use with order, equal units of measurement, and an absolute zero that indicates an absence of the variable being measure. Time: 0= no elapsed time |
|
|
Raw Score |
a data point that has not been transformed, analyzed, or changed in any way |
|
|
Repeated Measures |
Research in which one group of people is measured repeatedly or is in each condition of the study. SPSS calls the repeated measures IV within-subjects factor. -One-way ANOVA= repeated measures -Factorial ANOVA= Mixed Design |
|
|
Replication |
The duplication of scientific results, ideally in a different context or with a sample that has different characteristics. We estimate duplication with P-rep |
|
|
Research Hypothesis |
AKA: Alternative Hypothesis -This is the hypothesis we hope to support in our research predicting that a significant difference or relation exists between the groups we are comparing. -Usually opposite of the null |
|
|
Sample |
The group of people who participate in a study. -N=20 |
|
|
Sampling Distribution |
-a distribution of sample means based on random samples form a population. -The Central Limit Theorem says this distribution is almost always normal, no matter what the population is like |
|
|
Sampling Error |
The amount of inaccuracy in data due to studying a sample rather than testing the entire population. - can be due to a variety of extraneous variables and we can estimate it by what remains after we've determined the strength of our IV |
|
|
Single-Group Design |
A research method which measures only one group of participants. -Use a one sample t-test |
|
|
Source Table |
A table showing the important values, calculations, and final results of an ANOVA in an easy to read format. |
|
|
Standard Deviation |
A measure of variation, the square root of the average squared deviation from the mean. -Most commonly used measure of variation in psychology and along with variance (SD squared) is an integral part of many inferential statistics |
|
|
Standard Error of the Mean |
The SD of the sampling distribution, which we can approximate using statistics from samples. |
|
|
Standard Normal Distribution |
A normal distribution with a mean of 0 and a SD of 1. When raw scores are converted to this distribution they are called standard scores or z-scores. We use z-scores for comparing scores from different distributions and primarily for calculating percentiles. |
|
|
Standard Scores |
A raw score is expressed on a standardized scale relative to the mean and standard deviation. - Raw scores from several different measures can be converted to standard scores so they can be compared with one another. -Common: z, T, and stanines |
|
|
Statistical Power |
The probability of being correct when we reject the null hypothesis |
|
|
Statistical Significance |
An observed difference or relation between tow descriptive measures that is unlikely to have occurred by chance. |
|
|
Subject Variable |
A subject variable is a characteristic of the people we are studying, such as their gender or age, which allows difference conclusions (quasi-experiment). |
|
|
Situation Variable |
A variable we can manipulate, which allows cause and effect conclusions -True Experiment |
|
|
Sum of Squares |
-the sum of the squared deviations of each score from some mean. - |
|
|
t distribution |
a set of distributions that, although symmetrical and bell shaped, are not normally distributed. |
|
|
t-Test |
A parametric inferential statistical test comparing one mean to another. -compares a sample means -i.e., IV vs DV -developed by William Gossett |
|
|
Third Quartile |
The 75th Percentile of data set and the top of the box in a box plot |
|
|
True Experiment |
A research design in which the IV can be manipulated and participants randomly assigned. |
|
|
Two-Tailed Hypothesis |
A research hypothesis in which we predict that the groups being compared are related or different, without specifying the direction. We will always do two-tailed tests and use the p< .05 level of significance because this is the most common procedure in psychology- we'll try to follow it as a matter of routine. |
|
|
Type I Error |
An error hypothesis testing in which we reject the null hypothesis when it is true. This is the most "serious" inferential mistake we can make and we are constantly mindful of it because we close each significant interpretation with p< .05, the likelihood we've made a Type I error. |
|
|
Type II error |
an error in hypothesis testing in which we retain the null hypothesis when it is false. This is the error we might commit when saying an outcome is not significant. The likelihood is beta, but we don't usually compute or report it, we just end the citation with N.S. |
|
|
Ubiquity of the Normal Curve |
The concept that the bell-shaped curve describes the approximate shape of the distributions of a surprising number of characteristics that vary. This concept is particularly useful in psychology where we study lots of kinds of behaviors, measure those behaviors in a variety of ways, and rely on parametric statistical procedures to analyze those measures. |
|
|
Variable |
a measurement or behavior that has at least two values- so it can vary- be variable. |
|
|
Variance |
the average of the squared deviations from the mean- a measure of spread. |
|
|
Within-Groups ANOVA |
The sum of the squared deviations of each score form its group mean. This is also refereed to as the error sum of squares and it is often labeled error in the ANOVA source table. It is called error because it is used to calculate the variance due to all the random, chance factors that can influence the data (DV) and we refer to this as error to distinguish it from the effect of the IV on the data. |
|
|
Within-Groups Variance |
The variance within each group; an estimate of variance due to all individual differences in the population. In the ANOVA source table this is MS (mean squared deviation); that's variance. |
|
|
z score |
A number that expresses how far a raw score is away from the mean in standard deviations units. -AKA: standard score -Normal distribution- z=0 is at mean See LLHDQCard |
|
|
z Test |
A parametric inferential statistical test of the null hypothesis for a single sample where the population mean and variance are known. |
