Term
|
Definition
| A finite number of observations |
|
|
Term
|
Definition
| A qualitative, quantitative or physical measurement |
|
|
Term
|
Definition
| A variable whose outcome is determined as the result of a random experiment |
|
|
Term
|
Definition
| All possible elements belonging to a defined group |
|
|
Term
|
Definition
| A subset (selected proportion) of a defined group |
|
|
Term
|
Definition
| Qualatative measurements. Can have numbers but no math |
|
|
Term
|
Definition
| Ranked data. Numbered but no math involved |
|
|
Term
|
Definition
| Values where zero is arbitrary/non existant |
|
|
Term
|
Definition
| Values where zero is arbitrsry |
|
|
Term
|
Definition
| Data where zero is absolute (there can be an absence of the measure) |
|
|
Term
|
Definition
| Sample where every member of the population has an equal chance of being selected |
|
|
Term
|
Definition
| Divide population into at least two groups and randomly sample from each |
|
|
Term
|
Definition
| Sample every knth member of the population |
|
|
Term
|
Definition
| Divide population into 2 or more groups. Select a few groups and sample from them |
|
|
Term
|
Definition
| Select participants who are convenient for you to sample |
|
|
Term
|
Definition
| How values of data compare in terms of size and rank and how frequent the sizes occur |
|
|
Term
|
Definition
| Relates data to a mid/balancing point of the data set |
|
|
Term
| Qualitative measures of Central Tendency |
|
Definition
|
|
Term
| Quantitative measures of central tendency |
|
Definition
|
|
Term
|
Definition
| Multiply all values by eachother and take the nth root. Good when there are few data points over a wide range |
|
|
Term
|
Definition
| Sum of weight times score. All divided by sum of scores. Good when you know the data is influenced to a certain degree by a factor |
|
|
Term
|
Definition
| Spread of data in terms of distribution around a measure of central tendency |
|
|
Term
| Simplest form of central tendency and pros and cons |
|
Definition
| Range. Simple and easy to do but does not account for all data and very sensitive to extremes |
|
|
Term
|
Definition
| Standard deviation / mean x 100. Good when values have different magnitudes or units |
|
|
Term
|
Definition
|
|
Term
|
Definition
| (3(mean-median)) / s . Large negatives are negatively skewed (less than -1), large positives are positiviely skewed (greater than 1), between 1 and -1 is not skewed |
|
|
Term
|
Definition
| Measure of peakedness. Large negatives are flat, large positives are peaked |
|
|
Term
|
Definition
| Graph of the frequencies in a set (the accompanying graph to a frequency distribution) |
|
|
Term
| Rules of determining classes |
|
Definition
| They do not overlap, 5-12 classes, nature of class is determined by data (discrete or continuous) and you don't need data in every class |
|
|
Term
|
Definition
| range / number of classes (rounded) |
|
|
Term
|
Definition
| Summary table of quantitative data |
|
|
Term
| Relative frequency distribution |
|
Definition
| Divide the frequency by the total number frequencies |
|
|
Term
| Cumulative frequency distribution |
|
Definition
| Add up successive frequencies to existing |
|
|
Term
|
Definition
| GOod for seeing a trait across populations but not really used in stats |
|
|
Term
|
Definition
| Good to see change over time. Can be used to accompany stats |
|
|
Term
|
Definition
| Display relationships between 2 variables. Not really used in stats. But helpful in correlational/regresion |
|
|
Term
|
Definition
| The opposite of stats. We set a probability and used stats to test if we're right. Classic definition is event we're interested in divided by total number of events possible |
|
|
Term
| Relative Frequency Approximation |
|
Definition
| Probability = proportion of an event occuring over a large period of time |
|
|
Term
|
Definition
| The more samples, the closer the probability will be to the true probability |
|
|
Term
|
Definition
| Probability is based on knowledge of relevant circumstances (i.e. POP) |
|
|
Term
|
Definition
| Probability of A given B has occured |
|
|
Term
|
Definition
| One event does not affect the other |
|
|
Term
|
Definition
| One event does affect the other |
|
|
Term
| Mutually exclusive events |
|
Definition
|
|
Term
|
Definition
| If the probability of an event is known, its complementary event is every other possibility |
|
|
Term
|
Definition
| The total number of simple events that could occur in an experiment |
|
|
Term
|
Definition
| An activity with an uncertain outcome |
|
|
Term
|
Definition
| One or more possible outcomes of an experiment |
|
|
Term
|
Definition
| Only one event is possible |
|
|
Term
|
Definition
| A list of all possible outcomes in an experiment |
|
|
Term
|
Definition
| One that can take on any varialbe from a list. Its probability distribution is a list of all possible values and their frequencies |
|
|
Term
|
Definition
| Only two possible outcomes. Success is p and failure is q = 1-p. Trials are independent and p is constant. |
|
|
Term
|
Definition
| Only two possible outcomes. p and q=1-p. Trials are independent and p is constant. Applies where probability of success is rare |
|
|
Term
| Continuous Random Variable |
|
Definition
| Can take on an infinite number of values between 2 points |
|
|
Term
| Continuous Probability DIstributions |
|
Definition
| Can take on any shape, doesn`t give probability of individual events, probability is represented by area under curve |
|
|
Term
| Normal Gaussian Distirbution |
|
Definition
| Symmetrical, continuous, mean=median=mode |
|
|
Term
|
Definition
| Normalize curves. z = x minus mean divided by standard deviation |
|
|
Term
| Sampling distribution concept |
|
Definition
| Probability distribution of sample means with a sample mean of size n. Good to see if sample mean is close to population mean |
|
|
Term
| Assumptions of sampling distribution concept |
|
Definition
| If population is normally distributed sample is normally distributed, sample mean = population mean, sample stan dev is less than population |
|
|
Term
|
Definition
| Standard deviation divided by root of n |
|
|
Term
|
Definition
| Shape of distribution of mean will approximate normal curve if sample size is large enough (greater than 30) |
|
|
Term
| Conditions of Central Limit Theorum |
|
Definition
| Distribution of mean approaches normal distribution as n increases, mean of samples will equaly mean of population, standard deviation of sample means will approach standard deviation divided by the root of n. |
|
|
Term
| 7 Steps to Hypothesis testing |
|
Definition
1. Define H1 and H0
2. Chose significance level (alpha)represents chance of rejecting a true null. (Beta is chance of failing to reject a false)
3. Chose appropriate sample distribution and test stat (SND for large samples z = (x minus mean) divided by (standard deviation divided by root of n) Draw distribution tails (remember h0 is equality statement!)
4. Determine critical values using z-tables and draw them
5. Calculate a test stat
6.Compare values and make a decision
7. Write conclusion |
|
|
