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;
39 Cards in this Set
- Front
- Back
The probability that a sample was randomly drawn from population is measured by...
|
z test and T test. (diff with t test is that you know pop mean, but not pop strd dev)
|
|
Did incoming grad students score significantly better than the population who took the GRE?
This question is addressed by what type of test? |
z test ( 1 tailed)
|
|
Do students in Lubbock score significantly beeter or worse than students in TX on statewide achievement tests?
|
Z test
|
|
The t distribution approaches the z distribution as...
|
sample size approaches infinity
|
|
In T Testing, for a value to be considered significant, what must be true of the critical and observed t values?
|
t(observed) must be creater than t(critical) in order for something to be significant.
|
|
In terms of p values, what must be true for siginifcance?
|
For something to be signficant, p must be less than .05
|
|
Is performance better than chance in a 5 alternative test where chance =.20? Are corrlations between predictions and performance that I find for each participant significantly greater than 0?
These questions can be adressed by using what type of test? |
t test
|
|
What are the 3 types of t tests? Describe the differences between each type of test.
|
1.single sample t test
2. dependent (paired samples) t test 3. independent (2 sample) t test 1.determines the probability that the sample was drawn from the population (or not); want to decide how likely it is that your sample is a random sample from the population info given: pop mean, but not pop strd dev 2.-Testing whether the avg difference between two scores is zero. -compares knowledge at begining and end of course to see if knowledge incr 3.-compares two samples to determine whether they came from the same population or not -determines wether a difference between groups is likely/unlikely to have occured by chance info given: 2 means, but no info about pop mean or pop std dev |
|
determines the probability that the sample was drawn from the population (or not); want to decide how likely it is that your sample is a random sample from the population
|
single sample t test
|
|
.-Testing whether the avg difference between two scores is zero.
-compares knowledge at begining and end of course to see if knowledge incr |
dependent (paired samples) t test
|
|
compares two samples to determine whether they came from the same population or not
-determines wether a difference between groups is likely/unlikely to have occured by chance |
independent samples t test
|
|
info given: 2 means, but no info about pop mean or pop std dev
|
independent samples t test
|
|
info given: pop mean, but not pop strd dev
|
single sample t test
|
|
This type of test could help you examine whether males or females do better in PSY 5380
|
independent samples t test
|
|
Was the sample selected by student 1 different from the sample selected by student 2?
|
independent t test
|
|
Is the control group mean different from the treatment group mean (is it likely that both means were drawn randomly from the same population)?
|
independent t test
|
|
Do males and females differ in their response to an advertisement?
|
independent t test
|
|
Distinguish between the 4 types of scales.
|
nominal
ordinal interval ratio |
|
Classification in terns of mutually exlusive groups that do not have magnitude relationships.
|
nominal scale
|
|
breed of dog, gender
|
nominal scale
|
|
reflects differences in magnitude, but equal intervals cannot be assumed and their is no zero point
|
ordinal scale
|
|
national ranks of football teams, student's high school class rank
|
ordinal scale
|
|
reflects differences in magnitude and has equal intervals but no absolute zero point
|
interval scale
|
|
IQ scores, ACT/SAT scores, temperature scales
|
interval (be/c no true zero)
|
|
reflects magnitude, equal intervals, and has a zero point
|
ratio scale
|
|
dollars to euro conversion is reflective of which scale. Why?
|
ratio scale, because equal intervals with zero pt.
|
|
Parametric statistics are used with which scales?
|
interval and ratio scalses
|
|
nonparametric statistics are used with which scales?
|
nominal and ordinal
(we don't cover in this class) |
|
What is the probability that a mean that large or larger would
come from a population with Mu =5.68? |
single sample t test ( in this case, 1 tailed)
|
|
A person scored 600 on their GRE. The population mean for GRE is equal to 500, with a standard deviation of 100.
b) What percentage of the population had the same score or lower than this person? |
z score
|
|
I have a sample with N=9 and Y bar =5.11. My population has a known mean (Mu=5.68) and a known standard deviation (sigma = 2.24).
a) What is the probability of getting a Z this low or lower? b) What is the probability of getting a Z this high or higher? c) What is the probability of getting a z= 0.76 or bigger and z= -0.76 or lower |
Z test
|
|
A group of college students was randomly assigned to a lecture group or a World-Wide
Web (WWW) group. The lecture group heard a lecture on independent samples ttests. The WWW studied the same material as presented on the Web. Then both groups were given a test over the material. The results (percent correct on the test) were as follows: |
Independent Samples t Test ( 2 sample t test)
|
|
. Pre- and Post-test scores for students are provided. Did the students’ performance increase significantly as a result of taking the course?
|
Dependent t test (Paired Samples t Test)
|
|
. PSY 1300 students answer questions from practice GRE at the beginning of semester.
They answer 20 questions. Each question has 5 alternatives. a) Is the students’ performance significantly different from chance? b) Construct a 95% confidence interval around the sample mean. |
Single sample T test
|
|
I have a sample with a mean of 7.2 and N=9. The relevant population has a Mu = 5.68, Sj = 2.40.
a) Is my mean larger than would be expected if I had drawn my sample randomly? b) Is my mean larger OR smaller than would be expected if I had drawn my sample randomly? |
Single sample T test
|
|
Describe how Confidence intervals are calculated.
|
lower limit: sample mean - tcritical(standard error of mean)
Upper limit: sample mean + tcrit(standard error of mean) |
|
Describe the central limit theorem
|
If a population has a finite variance sigma squared an a mean mu, then the distribution of sample means from samples of n independent observations approaches a normal distribution with variance sigma squared/n and a mean mu as sample n increases
|
|
Large sample sizes tell you what about the variance of the sampling distribution.
|
Larger sample sizes mean less variance in the sampling distribution
|
|
The CLT tells us what about the sampling distribution of means
|
-sampling distrbution of means = population mean ( if thousands of random samples are taken)
-the sampling distribution of means will approach normality with larger samples, even if the population is not normal. |