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;
52 Cards in this Set
- Front
- Back
What are 2 types of statistics? |
Descriptive and inferential |
|
What is descriptive statistics? |
Procedures/tools to describe individual variables and relationships between variables; baseline demographic characteristics of population |
|
What tools are employed in descriptive statistics? |
Means, standard deviation, normal distribution, Z scores |
|
What are inferential statistics? What do they allow? |
Procedures used to analyze data after an experiment is completed and to determine if and independent variable has a significant effect. The making of extrapolations from a sample to greater population. |
|
What are 3 levels of measurement used in descriptive statistics? |
1. Nominal 2. Ordinal 3. Ratio |
|
What is nominal level measurement? What are 3 examples? |
No underlying continuum; arbitrary values assigned to discrete categories Religious affiliation, gender, political party |
|
What is ordinal level measurement? What is an example? |
Underlying continuum; values ordered but intervals are not equal; ranking system Community size (100-500 [1], 501-2000 [2], etc.) |
|
What is ratio level measurement? What are 3 examples? |
Underlying continuum and a zero value that aligns with true zero; equal intervals Weight, age in years, income |
|
What is central tendency? |
Simple statistics that typify a set of values |
|
What are the 3 measures of central tendency? |
1. Mean 2. Median 3. Mode |
|
What is a mean? How is it calculated? What level of measurement is used? |
Average Summing values and dividing by total number of cases Ratio level data |
|
What is a median? How is it calculated? What level of measurement is used? |
Middle most value; one half of cases above and other half below value Ordering set of values and taking middle number -if even no, divide middle two and take mean -if odd, take middle number Ratio and ordinal level data |
|
What is a mode? How is it calculated? What level of measurement is used? |
The most frequently occurring value Category of a variable with the most cases Ratio, ordinal and nominal level data |
|
What do measures of dispersion describe? What are the 3 types? |
The variability of measures; dispersion around mean 1. Range 2. Standard deviation 3. Variance |
|
What is range? How is it calculated? What is 1 example? |
Indicates the highest and lowest values Subtracting lowest from highest value Seniors in study range from 65 to 94; 29 year age range |
|
What is standard deviation? What level of measurement is used? What is it related to? If the sd is smaller, what does this indicate? |
Average amount of deviation from the mean value in a set of values Ratio level data Normal distribution curve Less variability in the dependent variable |
|
What is variance? How is it calculated? What level of measurement is used? |
A single number that represents the total amount of variation in a distribution Squaring the standard deviation Ratio level data |
|
What does standardizing the data allow? What is 1 example? |
Comparisons between units of different sizes; relative rank based on standardization Suicide rates in different provinces -9.90/100,000 ppl in Ontario -32.46/100,000 ppl in NWT |
|
What are the 5 types of standardized data? |
1. Proportions 2. Percentages 3. Percentage change 4. Rates 5. Ratios |
|
What are proportions? What do they range from? Can be used in place of what? |
A standard method of designating a portion of a total .00 to 1.00 Percentages |
|
What are percentages? How are they calculated? What do they represent? |
Denote probability by converting proportions to percentages Proportion x 100 = % How often something happens per 100 times |
|
What is percent change? How is it calculated? What type of study is it important for? What level of measurement? |
A measure of how much something has changed over a given time period (Time 2 - Time 1) / Time 1 x 100 RCTs Ratio level data |
|
What are rates? Incidence per____ or _____ What is 1 example? |
The frequency of something for a standardized sized unit 1000 or 10000 (can also be 100000) 104 suicides in a pop. of 757, 465 SR= 104/ 757, 465 x 100, 000 = 13.73 13.73 suicides per 100, 000 |
|
What are ratios? What is 1 example? |
Represent a comparison of one thing (rates) to another 200 burglaries per 100, 000 in the US compared to 57 per 100, 000 in Cad, the US/Cad ratio is: US/Cad = 200/57 = 3.51 US rate 3.5x higher than Cad |
|
What is normal distribution? What are 3 other terms for it? What concept is it based on? |
A symmetrical, bell-shaped distribution with half the scores above and half below the mean Normal curve, bell curve, Gaussian distribution As the number of scores in a sample increases, the scores approximate a more normal distribution (closer to the mean) |
|
Can have a ____ or ____ skewed distribution Can you perform stats on skewed data? What would have to be done in order to? |
Plus or minus No Transformation: mathematical redistribution; logarithm |
|
What 3 factors will remain the same in a normal distribution? How do you measure distances from the mean? |
Mean, median, mode Standard deviation units |
|
How many cases will be +/- 1 sd of the mean? +/- 2 sd? |
2/3 of cases 68.2% 95.6% |
|
What is a Z score? What is the formula? What is the range? |
A standardized score that represents the distance above or below the mean, in sd units, of any raw value in a distribution Z = [X (observation) - Xx (mean)]/ sd +3 to -3 |
|
If the sample mean income is $65K and the sd is $22K and a participant has a Z score of +1.43, how much do they earn? |
1.43 x $22, 000 = $31, 460 $31, 460 + $65, 000 = $96, 460 |
|
What are 3 means of analysis that one can use to describe relationships between variables? |
1. Crosstabular analysis 2. Comparison of means 3. Correlations |
|
What are the 3 steps in performing a data analysis? |
1. Decide IV and DV 2. Decide appropriate procedure for examining relationships 3. Perform analysis |
|
What is a crosstabular analysis? What level of measurement is the DV and IV? |
Data is sorted into categories within the IV and DV (cross-classified) to show the relationship between them DV: nominal or ordinal IV: nominal, ordinal or ratio |
|
What type of analysis is a crosstab? |
Bivariate analysis |
|
What is a contingency table and when is it used? |
Presents info so the relationship between a nominal level dependent variable can be related to an independent variable. Used in crosstabular analysis. |
|
What are the steps in performing a univariate analysis? |
Analyze > Descriptive statistics > Frequencies > Statistics > Charts |
|
What are the steps in performing a crosstab analysis? |
Univariate frequencies/charts > Analyze > Descriptive statistics > Crosstabs > DV rows, IV columns > Cells (%) in direction of IV |
|
What is comparing means? When it is used? |
Comparing the mean values of the dependent variable for each category of the independent variable When the dependent variable is ratio level data and when the independent variable is either nominal or ordinal |
|
What 2 tests of significance may be used in comparing means? |
T-test and ANOVA |
|
What does a Chi Square test? What does it provide? How do you know if something is statistically significant? |
The null hypothesis that 2 variables are independent A p value If the p value is <0.05, it is statistically significant, if the p value is >0.05, it is not statistically significant |
|
What are the steps in performing a Chi Square? |
Analyze > Descriptive Statistics > Crosstabs > DV rows, IV columns > Cells (%) > Statistics box > Chi Square test > Clustered bar charts |
|
What is a correlation? When is it used? Can it be used in inferring cause and effect? |
Describes how closely 2 ratio level variables co-vary together; associations Ratio level variables No |
|
What are the 2 basic questions that are asked of the correlation? |
1. What is the equation that describes the relationship between the 2 variables? 2. What is the strength of the relation between the 2 variables? |
|
How does one determine the strength of a correlation between 2 variables? |
R value (correlation coefficient) -ranges from -1 to +1 -over 0.9 is a strong correlation -0.5 is moderate |
|
What is r = -1.0? What is r = +1.0? What is r squared? |
-1.0 is a perfect negative correlation; an increase of 1 unit in 1 variable is associated with a proportional decrease in the other variable; negative slope +1.0 is a perfect positive correlation; positive slope The amount of variance in the X axis that explains the effect on the Y axis R 0.9(0.9) = 0.81 = 81% |
|
On which axis are the independent and dependent variables? What is the general equation for the correlation slope? What does it estimate? |
Independent variable on the X axis Dependent variable on the Y axis Y=a + bX How much the IV has to change in order to produce a unit of change in the DV |
|
What type of correlation should be used for ratio level variables? Ordinal level? |
Ratio- Pearson Spearman- Ordinal |
|
What is meant by intraclass correlation coefficient? |
Describes how strongly units in the same group resemble each other (categorical data) |
|
What happens on the Y axis if the r value is 1? |
Means 100% increment; every unit increase on the X axis will be the same unit increase on the Y axis |
|
What is a regression line? |
Line that offers the best linear description of the relation between the 2 variables |
|
In the equation Y= a + bX, what does a and b mean? How is b derived? |
A is the constant or the point at which the line crosses the X axis B is the slope and it represents height (h)/ base (b) |
|
What can you predict using the equation? |
The value of the dependent variable given the value of the independent variable |