Proportion = Frequency x 100 = Percentage Total No | Z score (standardised value)-how many sds from the mean the value liesZ score = data value – mean Standard deviation | Metric Data = ExploreCategory = Frequencies| Bigger sample size will give a narrower confidence interval range (more specific) outliers affect the mean but not the median – this is why the median is preferred here.mean| | Reports -Only give confidence interval if significant-All values to 2 dec pts except the p-value Experimental = IV is manipulated to see the effect on the DV Observational = Information just observed & recorded|
P-Value Significant Figurep-value < 0.05 = Significantp-value < 0.05 = Not SignificantReport p value 0.000 as <0.001 The probability that our test statistic takes the observed value Always leave at 3 decimal places| Levene’s Test-Used to test if equal variancesIf significant (<0.05)– use equal variances not assumed rowIf not significant (>0.05)– use equal variances assumed rowReport confidence interval as the 95% confidence interval indicates...| Dependent Variable = the variable in which we expect to see a changeIndependent Variable = The variable which we expect to have an effect on the dependent variable Example: There will be a statistically significant difference in graduation rates of at-risk high-school seniors who participate in an intensive study program as opposed to at-risk high-school seniors who do not participate in the intensive study program." (LaFountain & Bartos, 2002, p. 57)IV: Participation in intensive study program. DV: Graduation rates.| Nuisance variable
- Associated with the DV
- Never the IV-Must Vary| Confounding Factor = a nuisance variable associated with both the DV & the IV -Alters the logic of the experiment (cannot tell if it is the IV or the nuisance variable that is having an effect on the DV| Control by Nuisance variables & confounding factors by Randomisation & couter-balancing| Scatterplots Coefficient
DV on Y AxisIV on X AxisRegression Equation = Y = a + bx
Used to predict the value of the Dependent variable if we know something about the Independent variable Y = DV, a = vertical intercept (constant), b = slope Co-efficient of determination (r squared) (standardized corfficents)Interpretation: 18% of the variation in revenue [DV] can be explained by the linear relationship between the distance from city centre [IV] and revenue | | Correlation Coefficient (Pearson’s r)-Measures the strength of the linear relationship.8 or more = Strong.5 to .7 = Moderate.3 to .4 = WeakLess than .2 = Extremely Weak-How important is the relationshipr = Pearson’s r X Pearson’s r|
IDENTIFY HOW VARIABLES ARE MEASURED (METRIC/CATEGORICAL), THEN WHAT WE ARE COMPARING? Comparing means – Comparing our sample to a population already known – 1 Metric Variable - Explore- ONE SAMPLE T-TEST Comparing means (Ours is bigger than yours) – Comparing two independent groups – 2 Variables = 1 Metric & 1 Category – INDEPENDENT SAMPLES T-TEST Comparing means (Our this is bigger than our that) – Comparing 2 sets of data for the same group (same people but different conditions) (repeated measures) – 2 Variables = 1 Metric & 1 Category - PAIRED SAMPLES T-TEST Comparing a proportion/Percentage –Comparing our sample to a population already known – 1 Categorical Variable – Frequencies - BINOMIAL TEST Looking for a relationship (More of this means more of that) – 2 Metric variables - CORRELATION/REGRESSION (Pearson's r) Looking for a relationship (We’re more likely than you are) – 2 Categorical variables – CROSSTABULATIONS/CHI-SQUARE One Sample t-test Report Writing
A study was conducted to investigate whether the age of tourists in Katatonia has increased since 1995. In a random sample of 250 tourists in Katatonia, the average age was 47.77 years (s = 15.41 years). This is higher than...