Research Exam 3(Chpt 14)
what number is the desired α level of significance?
.05
three types of t-test
1. one-sample t-test 2. dependent or paired-samples t-test 3. independent t-test
issues with t-test
1. whether or not the researcher used the correct t-test for the experiment 2. probability of committing a type I error
Two branches of statistics?
Descriptive and inerential
T or F: The lower the α level, the HIGHER the probability that the researcher will commit a Type I error
False --> the lower the probability
T or F: Power is the probability of rejecting the null hypothesis when it is indeed false or the probability of accepting the research hypothesis
TRUE
T or F: The STRONGER the power a study has, the LESS of a chance of committing a Type II Error
TRUE
T or F: inferential stats allow us to make generalizations that go beyond our sample?
TRUE
T or F: the LOWER the power a study has, the GREATER the chance of committing a Type II Error
TRUE
T or F: the lower the p-value , the LESS LIKELY the obtained differences in the results are due to chance and the MORE LIKELY the obtained differences in the results are due to manipulation of the study
TRUE
T or FL obtaining statistically significant results might not be the end goal after all → ideal combination would be to find significant results that contribute to the field in a meaningful way
TRUE
inferential statistics
allows you to draw conclusions about the population of interest from the sample used
mean, median and mode in a skewed distribution?
are not the same
mean, median and mode in a normal distribution?
are the same
when does one determine the level of significance?
before conducting the experiment
Bonferroni Correction
correction used when conducting multiple statistical tests to limit the possibility of conducting a type I error → take .05 and divide by the number of t-test you plan on conducting
Descriptive statistics
describe the data
independent t-test
evaluates whether or not means from two different groups are significantly different from one another
if the p-value obtained from a study is more than the conventional alpha level of .05 then we..?
fail to reject the null hypothesis
Type II Error
failing to reject the null hypothesis even though the null hypothesis is false
what is a frequent was of summarizing a data set?
giving measures of central tendency
Normal Distribution
if all three measures (mean, median and mode) are similar
What does knowing the amount of standard deviation help determine?
if the mean is a representative score for the data set
what do the results of an ANOVA study tell you?
if there are differences between the levels,and follow-up analyses will tell you where the particular differences occurred
failing to the reject the null hypothesis
indicating that the means of the groups are equal (no difference)
rejecting the null hypothesis
indicating that there is a statistically significant difference between the means of the groups
what does a study replication do?
it verifies results
one way researchers can reduce the chance of committing a type I error when conducting multiple t-test
making the alpha level more stringent (Bonferroni Correction)→ take .05 and divide by the number of t-test you plan on conducting
What are the three commonly used measures of central tendency?
mean, median and mode
what measure does one use in a skewed distribution to best represent the date set?
median
what does a SMALL standard deviation indicate?
most scores are located around the mean
what are measures of central tendency?
numbers that represent the scores in data
level of significance (aka alpha level: α)
probability of obtaining data assuming the null hypothesis is true
variance
provides information on the spread of scores in a set
descriptive stats summarize data by
providing numbers which are representative or typical for the set
If the p-value obtained from a study is less than the conventional alpha level of .05 then we...?
reject the null hypothesis
two options researchers have when testing a hypothesis
reject the null hypothesis or fail to reject the null hypothesis
Type I Error
rejecting the null hypothesis when the null hypothesis is in fact TRUE
What is one of the best ways to ensure a result is accurate and not a product of an error?
replicating the study
If a replicated study's results are the same as the original then...?
researchers can be confident a decision error was not committed
How can one determine if the groups tested are statistically different from one another?
researchers perform statistical procedures and examine the level of significance
What additional information should be provided when conducting inferential stats (according to the APA)
sample size of each condition, means and standard deviations for each condition, the obtained` statistical value, degrees of freedom, the obtained p-value, effect size, and confidence intervals
Where do the scores fall in normal distribution?
scores cluster around the middle and few, but equal, #s cluster on the extremes
what does a LARGE standard deviation indicate?
scores vary widely from the mean and are spread out in the data set
a distribution that is NOT normal is called
skewed distribution
Two types of inferential statistics
t-test (most common) and ANOVA
If a replicated study's results were not the same as the original then...?
the conclusions from the original study would need to be reevaluated
standard deviation
used most frequetnly to provide info on how much scores vary from the mean
dependent or paired-samples t-test
used to determine if the mean difference between the pairs is significantly from zero
one-sample t-test
used to determine when a mean is significantly different from a constant
Analysis of Variance(ANOVA)
used when you have an independent variable with 3 or more levels OR when you have multiple independent variables
two measures of variability that can also be used to summarize data
variance and standard deviation
when does a Type I error occur?
when a researcher finds the groups to be difference when they are really the same
when does the replication of a study occur?
when another researcher/same researcher recreates a study using the same methodology as in the original study
Skewed Distribution
when most frequently occurring scores are not in the middle → scores pile up at one end of the distribution or the other end
When does a Type II Error occur?
when the researcher finds the groups to be the same when they are really different