Term
|
Definition
| Data-Based Decision Making - Used in schools to help describe problems, direct activities and resource allocations. |
|
|
Term
|
Definition
| Ambitious but attainable goals/benchmarks. |
|
|
Term
| Why can evidence be misleading? |
|
Definition
| Due to bias, preconceptions, assumptions, how we see the data, multiple perspectives. |
|
|
Term
| What are some examples of disaggregated data? |
|
Definition
| Gender, race, grade, SES, language, special education, etc. |
|
|
Term
| If data suggest changes or differences, how do you know if the changes are meaningful? |
|
Definition
| Variability (ie: large enough sample size, enough data). |
|
|
Term
| Statistics provide a way of...? |
|
Definition
| Classifying, organizing, and analyzing data. |
|
|
Term
| What's the difference between descriptive and inferential statistics? |
|
Definition
Descriptive stats organizes and summarizes data (ie: How many? Average scores).
Inferential stats tells us what data means (ie: Is there a correlation, is it due to probability/chance?). |
|
|
Term
| What is the difference between a population and a sample? |
|
Definition
Population is ALL people, places or things in question (ie: ALL middle school students in US).
Samples is a subset of the populations (ie: SOME of the middle schools students in US, middle school students in PA). |
|
|
Term
| What is the difference between a statistic and a parameter? |
|
Definition
Statistic speaks of a sample.
A parameter speaks of a population. |
|
|
Term
| Why should a sample be representative of a population? |
|
Definition
| For accuracy and validity. |
|
|
Term
|
Definition
| Sample selection where each member of the population has an equal chance to participate. |
|
|
Term
|
Definition
| A characteristic that can change/vary (ie: teacher, gender, age). |
|
|
Term
| What is the difference between an independent and dependent variable? |
|
Definition
Independent variables are the predictor and can influence the outcome (ie: gender, race, age).
Dependent variables are the outcome (ie: test scores, depression, reading level). |
|
|
Term
| What is an extraneous variable? |
|
Definition
| Any factor (other than the independent variables) that can affect the outcome/dependent variable. |
|
|
Term
| What is a control variable? |
|
Definition
| Extraneous variables that are used to control effects/results (ie: only selecting those from a certain variable - age, SES). |
|
|
Term
| What is a nominal scale of measurement? |
|
Definition
| Categorical data, # has no value, descriptive, to differentiate between categories, the least complex scale. |
|
|
Term
| What is an ordinal measurement scale? |
|
Definition
| Based on ranking or order, lichert scales. |
|
|
Term
| What is an interval measurement scale? |
|
Definition
| Numbers with equal units of measurement (ie: test scores, grades). No absolute zero. |
|
|
Term
| What is a ratio measurement scale? |
|
Definition
| Number with equal units of measurement but WITH an absolute zero (ie: weight, height). |
|
|
Term
| What is a frequency distribution? |
|
Definition
| Organizes ungrouped data to show the number of times each score in a data set was obtained. |
|
|
Term
| What is a bell-shaped distribution? |
|
Definition
| Normal distribution or normal curve. Two tailed distribution w/each tail showing low frequencies of extreme scores. |
|
|
Term
| What is a skewed distribution? |
|
Definition
| Asymmetrical distribution. Have only one pronounced tail. Can have positive or negative skewed distribution. |
|
|
Term
| What is a rectangular distribution? |
|
Definition
| No variability. Everyone got the same score. |
|
|
Term
| What are the 3 measures of central tendency? |
|
Definition
| Mean - average. Median - middle. Mode - most frequent. |
|
|
Term
| What are some problems with the mean? |
|
Definition
| Not accurate for highly skewed data. Highly effected by extreme scores - then should use median. |
|
|
Term
| What are some problems with the median? |
|
Definition
| Ignores a lot of info from data set. Not the first preferred way of describing data. |
|
|
Term
|
Definition
| The most frequent value. Most useful for nominal data. Can have more than one mode (bimodal). Can not have a data set without a mode. Normal distribution has one hump so is unimodal. Not very stable. |
|
|
Term
| What are the 3 measures of variability? |
|
Definition
| Range, variance, standard deviation. |
|
|
Term
|
Definition
| Simplest way to measure variability. Highest minus lowest score. |
|
|
Term
|
Definition
| S squared. How close the scores are to the center of the distribution. The mean of the squared differences of the scores from the mean/The average of the squared deviation scores. |
|
|
Term
| What is a deviation score? |
|
Definition
| How many points a score is above or below the mean. Score minus the mean. It shows variability. |
|
|
Term
| What is the average of all the deviation scores in a distribution? |
|
Definition
|
|
Term
| How do you calculate variance? |
|
Definition
| The sum of the (scores minus the mean) squared, divided by n. |
|
|
Term
| What is the standard deviation? |
|
Definition
| Indication of how different a score is from the mean. The most widely used measure of variability. The square root of the variance. Best for normal curves. |
|
|
Term
| What is associational/relational research? |
|
Definition
| Used to look at the correlation. |
|
|
Term
| What is comparative research? |
|
Definition
| Used to look for statistically significant differences. Quasi-experimental (not a random sample). Experimental (random sample, causality). |
|
|
Term
| What is chi-square statistics? |
|
Definition
| To find relations between TWO nominal/ordinal categories (ie: is SES related to career choice?). |
|
|
Term
| What is a t-test and ANOVA? |
|
Definition
| Tests for statistically significant differences between 2 interval/ratio groups. ANOVA tests between more than 2 groups. |
|
|
Term
| What is the general linear model? |
|
Definition
| There is a relationship between the dependent and independent variables. Creates a regression line using data points. |
|
|
Term
|
Definition
| The distance from each data point to the straight regression line. |
|
|
Term
| What is the difference between high and low variability? |
|
Definition
| High variability has a lot of overlap amongst the scores and a large spread distribution. |
|
|
Term
| What is a paired sample t-test? |
|
Definition
| Compares the mean of one group at 2 different points in time. The independent variable is time and the dependent variable is the score/outcome. |
|
|
Term
| What is an independent sample t-test? |
|
Definition
| Compares the mean of TWO groups to each other. The independent variable is gender for example, and the dependent variable is the score/outcome. |
|
|
Term
| What is a non-directional/two-tailed hypothesis? |
|
Definition
It asks if there's any statistically significant difference between two groups (ie: is there a difference in math grades between boys and girls?).
Null hypothesis (H o) - No difference
Alternative hypothesis (H a) - There IS a difference. |
|
|
Term
| What is a directional/one-tailed hypothesis? |
|
Definition
| It asks is one greater/higher than the other? (ie: Are boy's math scores significantly higher than girls?). |
|
|
Term
| What is a one-way ANOVA/analysis of variance? |
|
Definition
| Used to compare the means of 3 or more groups of the same size (ie: advanced, regular, remedial). |
|
|
Term
| What is a repeated ANOVA measure? |
|
Definition
| Used to compare the means of ONE group at 3 or more points in time (ie: grades at different marking periods). |
|
|
Term
| What is the difference between a criterion test and a norm-referenced test? |
|
Definition
Criterion test emphasizes mastery of a subject.
Norm-referenced tests emphasizes comparison. Can be for placement or awards. |
|
|
Term
| What is a normal distribution? |
|
Definition
| Scores that are distributed evenly around the mean. Mean, median, and mode are all the same. Social sciences typically have normal distributions (ie: IQ scores). |
|
|
Term
| What is the purpose of a standard score? |
|
Definition
| To describe the location of every score relative to the mean. |
|
|
Term
| What is the z-score and how do you calculate it? |
|
Definition
How many standard deviations a score is from the mean.
z = x - M / SD
x = (z) (SD) + M |
|
|
Term
| What is the mean and the standard deviation when you convert scores into a z-score? |
|
Definition
| The mean will be zero and the standard deviation will be 1. |
|
|
Term
| Does converting your raw scores into z-scores change the shape of the distribution of your data? |
|
Definition
| No, whether its a normal or skewed distribution, it stays the same. |
|
|
Term
| Why are standard scores like the z-score useful? |
|
Definition
| They give you a way to standardize and compare. |
|
|
Term
| What are the downsides of standard scores like the z-score? |
|
Definition
| Sometimes raw data is more meaningful. Not everyone can interpret it. |
|
|
Term
| What are some other common standard scores? |
|
Definition
| Z-score, T-score, NCE normal equivalent score, Stanine score, IQ score, SAT score. |
|
|
Term
| What is hypothesis testing? |
|
Definition
| Predicts outcomes and relationships between variables, finds probability. Tests which of two hypotheses is best supported by data (null or alternative-directional or non directional). |
|
|
Term
|
Definition
| The smallest level of significance where you would reject the null hypothesis. The null hypothesis is either accepted or rejected based on the determined p-value. This is used so you can know what the chances are of an error happening if we reject the null hypothesis. |
|
|
Term
|
Definition
| Rejecting the null hypothesis when its true. Claiming that there's a difference between 2 groups when there isn't. Alpha. |
|
|
Term
|
Definition
| The probability of making a type I error. |
|
|
Term
|
Definition
| Accepting the null hypothesis when its false. Claiming theres no difference between 2 groups when there is. Beta. |
|
|
Term
|
Definition
| The probability of making a type II error. |
|
|
Term
| What is a correlational and casual statistical relationship? |
|
Definition
Correlational relationships show similarities between variables (r=strength of correlation=closer to +1 or -1, the stronger the correlation, r=0 no correlation).
Casual relationship tells us that a change in one variable causes change in the other variable. |
|
|
Term
| Correlation can be effected if....? |
|
Definition
Data is truncated (limited range of values).
Data comparing extremely different groups causing overestimation of correlation.
Too small sample size. |
|
|
Term
| Will you get an A on this exam? |
|
Definition
|
|
