Home / Summaries / Class notes - Statistics 1 and 2 / variables-statistics-scores

Book page 1-5

36 important questions on Book page 1-5

What can be the cause of unlucky randomization?

Random assignment

Explain a typical non experimental research

A researcher measures two or more variables that are believed to be meaningfully related and the researcher does not introduce a treatment

What is one requirement for causal inference

The variable thought to be happen earlier in time than the variables though to be the outcome

What can satisfy the results from the non experimental studies?

The list of conditions for causal inference

When do we prefer to say that X and Y are correlated?

When there is not temporal precedence and no ability to rule out confounding variables

How are studies called that compare group outcomes but lack the full set of controls in true experiments?

Quasi-experiments

When do Quasi-experiments often arise?

When programs are evaluated in field settings

What are three major types of errors in interpretation?

1. When the researcher describes a association between variables but there is not enough evidence of the lack of confounding variables
2. Overgeneralizing
3. Misunderstanding or minimizing the limitations of the research design

What can be used to summarize information about the quantitative variables using simple descriptive statistics?

Mode, median and mean

How do we obtain information about variation of scores?

By examine minimum and maximum values, range, variance and standard deviation

How can additional information about central tendency and variability be obtained?

By computing descriptive statistics on the basis of the values of all scores

When does this question arise? Which one of two of these statistics do the best job of communicating information about typical or ''average'' values?

When the mean, mode and median differ

How to obtain the sample median for a set of N scores in a small data set?

1. Arrange the X scores in rank order from high to low
2. Determine the number of scores, c, that correspond to half of N.
3. Count the number of scores and see if N is odd or even

What are the three reasons behind using samples?

1. Obtain knowledge
2. Observing a population is feasible
3. We observe a part

What determines level measurement?

The kind of statistics and statistical analyses you can use meaningfully

What tells a test statistic us?

How far away our sample result is from the expected value under h0, taking into account sampling fluctuations

What are the 5 steps in a one-tailed test?

1a. Formulate hypothesis
1b. Set alpha
2a. Compute t statistic
2b. Compute the p value
3. Draw a conclusion

What are the assumptions that the Y (dependent) variable should satisfy?

1. Quantitative
2. They are independent of each other between and within groups
3. They are sampled from normally distributed populations with equal variances

What do we need to verify to determine about significance?

Whether our sample mean belongs to the alpha*100% most extreme sample means under H0

What are the two methods of determining significance?

1. Check whether the p-value is below alpha
2. Use alpha*100% and see if it falls in the reject region

What are the 5 steps of an independent t-test?

1. Formulate H0 and H1
2. Compute sample statistic
3. Compute test statistics and p-value
4. Determine if the test was significant
5. Draw conclusions

What means standard error of mean difference?

Variability in the mean difference between two samples if you would repeatedly draw samples of size n1 and n2 from the same populations

What happens if assumptions are violated?

1. We can no longer trust the results (and the p-values)
2. The estimated effects may not be meaningful

What does determine the best fitting straight line mean?

Determine the values of coefficients (b0 and b1) that predicts Y from X the best

What is the best filling line?

The one that minimizes the sum of squared errors (least-squares solution)

Interpretation unstandardized regression coefficient b...

When X increases with 1 unit, then Y increases with b units

Interpretation standardized regression coefficient b....

When X increases with 1 standard deviation, then Y increases with b standard deviation

What is our ultimate goal in science?

To find causal relationships

With multiple regression we can control for confounders by studying what?

Partial effect

What are 4 types of multivariate relationships?

1. Spuriousness association between X and Y
2. Mediation effect
3. Multiple causes
4. Interaction

What is the goal of multiple regression?

To test hypothesis about the influence of several independent variables on one quantitative dependent variables

Why are multiple independent variables useful?

1. Better understanding of theory
2. Controlling for third variable

For what two things do we use nested models?

1. Hierarchical regression analysis
2. Testing the partial effects of categorical predictors with three or more categories

What provides a dummy or dichotomous predictor variable?

Yes/no information for questions about group membership

When can a predictor become non-significant?

When another variable is statistically controlled

When is a two-dimensional graph sufficient?

For bivariate regression

The question on the page originate from the summary of the following study material:

Statistics 1 and 2

View summary

A unique study and practice tool
Never study anything twice again
Get the grades you hope for
100% sure, 100% understanding

Remember faster, study better. Scientifically proven.