Term
| What are alternatives to quantitative studies? |
|
Definition
Qualitative: Key informant interviews, focus groups, participant observation, etc
Mixed Method Studies: Iterative process, combining both Quan. and Qual. research methods |
|
|
Term
| The Mean is a measure of... |
|
Definition
central tendancy
- it's an AVERAGE |
|
|
Term
| Define the central limit theorem |
|
Definition
| When the average of a series of means is normally distributed, with a variance of s^2, a std deviation of s, and mean of u. |
|
|
Term
| Which test do we use for all of our means? |
|
Definition
|
|
Term
| What are the limitations of the CLT? |
|
Definition
1. Need large sample size
2. Need to know std dev of pop'n |
|
|
Term
| When do we need standard error of the mean and how do we get it? |
|
Definition
| - used with standard deviation of the sample when standard deviation of the population is unknown (which is often is) |
|
|
Term
| When do we use the t-statistic? (2 reasons) |
|
Definition
1. To test if a single mean is significantly different from zero or;
2. to test if two means are significantly different from one another |
|
|
Term
|
Definition
| A type of ratio where the value of the numerator is included in the value of the denominator |
|
|
Term
| Which ways can a proportion be expressed? |
|
Definition
Decimals (0.3), Vulgar functions (1/4), and percentages (20%)
They can only be expressed in a range from 0.0 to 1.0 (or 0% to 100%) |
|
|
Term
|
Definition
| A fraction measure where the numberator is NOT included in the denominator |
|
|
Term
| What is the proportion of fish in the sea if the water to fish ratio is 6:4? |
|
Definition
| 4/(4+6) = 0.4 = 40% = 4/10 |
|
|
Term
| Which test do we use to test the significance of proportions/ratios, and why? |
|
Definition
| X^2, because proportions/ratios cannot be negative, so we can't use the t-distribution |
|
|
Term
|
Definition
- The probability that the observed differences (b/w groups) might have arisen due to chance alone
- the probability that a test statistic would be larger than observed (if the null was true) |
|
|
Term
| What is the null hypothesis? |
|
Definition
That there is no difference b/w the two groups
or
That there is no association b/w exposure and outcome
or
that there is no difference b/w a measure and zero |
|
|
Term
| With regards to the null, how can we define the p-value? |
|
Definition
| The p-value is the probability of getting the test statistic that we calculated (or a bigger one), IF THE NULL IS TRUE. |
|
|
Term
| What does a high p-value mean? |
|
Definition
| That the difference we observed was due to chance. |
|
|
Term
| If p-value is high, should we accept the null hypothesis, or reject it and why? |
|
Definition
Null = there is no association/difference
High p-value = difference due to chance
So, we should accept the null, that there is no association/difference |
|
|
Term
| What does a low p-value mean? |
|
Definition
| Less likely that the observed difference was due to chance - means that there IS a difference/association |
|
|
Term
| If p-value is low, should we accept the null hypothesis, or reject it and why? |
|
Definition
| Reject the null, but a low p-value means that it's less like that the observed difference was due to chance, so it's more likely that, in reality, there IS a difference. |
|
|
Term
|
Definition
The rule to decide if the difference observed is due to chance (not significant), or not due to chance (significant).
0.05 (5%) used as industry std |
|
|
Term
|
Definition
|
|
Term
|
Definition
|
|
Term
| Define confidence interval |
|
Definition
| Reasonable statement about the entire population, based on sample statistics |
|
|
Term
|
Definition
| Out of 100 samples, we expect 95 of these to fall with in the bounds of the CI (the area in which we expect a certain result) |
|
|
Term
| A 95% confidence interval (CI) of 1.3 - 4.2 means that... |
|
Definition
| We can be 95% confident that the observed mean would fall b/w 1.3 and 4.2 |
|
|
