UNDERSTANDING STATISTICS
Wide bell curve?
-Large SD -More variability
Ratio
-difference can be quantitatively measured -absolute zero
Interval
-difference can be quantitatively measured -no absolute zero
Nominal
-naming level -participants will respond with words not numbers
Ordinal
-ordering from high to low -does not measure amount of variation between -can be quantified
Univariate Analysis
Analyzing how participants vary on only one variable
Bivariate Analysis
Analyzing the relationship between two nominal variables
Median score has ____% above and below
50%
In a normal distribution, _____ percent of the participants in the distribution lie within one SD off the mean
68%
Why do we use null hypoth?
Hypothesis: the loss of my socks is due to alien burglary. However, when you get your results, it's possible that any relationship that appears in your data was produced by random chance. In order to back up your hypothesis you need to compare the results against the opposite situation: that the loss of socks is not due to alien burglary. This is your null hypothesis - the assertion that the things you were testing (i.e. rates of alien activity and sock loss) are not related and your results are the product of random chance events.
Which is more reliable statistic? Range or IQR?
IQR
Objective, multiple choice achievement tests are usually used to measure at what level?
Interval
Weight measured in points uses what scale of measurement?
Interval
SD size of short and wide bell curve?
Large
Which group has less variability? Group with SD of 14 or group with SD of 20
SD of 14
Use mean or median ordinal measurement?
Median
IQR
Middle two quarters, or middle 50% of the participants
Bell Curve: Skewed to the right
Positive Skew
Measure of variability used when the MEAN measure of central tendency
Standard deviation
How do we determine the probability that the null hypothesis is true? (and consequently that our results are not)
Significant tests
T/F: If researchers do not sample, they do need inferential statistics
True Why? Because without a sample, it is sure that the study population is free of "sampling errors"
T/F: Null hypothesis can never be proven
True! Researcher can only reject or failed to reject based on probability of truth
When all values are the same, what is the SD?
zero
Frequencies and statistics are used with what branch of statistics?
descriptive
Null hypoth more likely to be stated in journal article or dissertation?
dissertation
Direct relationship
high on one variable are high on another
Inverse relationship
high on one variable are low on another
Do researchers reject null hypothesis when the probability of its truth is high or low?
low
Inferential Statistics
make inferences from samples to the entire populations from which the samples were drawn
Margin of error is used with what branch of statistic?
margin of error
Known as the balance point in a distribution
mean
Synonym for averages
measures of central tendencies
Used when distribution is highly skewed?
median
Most frequently occurring score
mode
Which of the three averages is seldom used in formal reports of research?
mode
Which scale of measurement is the lowest?
nominal
Which are harder to comprehend: proportions or percentages?
proportions
Does the term sampling error refer to random errors or bias?
random errors
Values obtained from a sample?
statistics
Descriptive Statistics
summarize and describe data
Alternative way to saying a researcher has rejected the null hypothesis?
the difference (of your research results) is statistically significant
Measure of variability used when the MEDIAN measure of central tendency
Range/IQR
Number of minutes of overtime work uses what scale of measurement?
Ratio
What does the null hypothesis say about sampling error?
The observed difference in results was created by random errors
T/F: The 68% rule of thumb does not strictly apply if the distribution is not normal
True
Which branch helps researchers to make generalizations about the characteristics of populations on the basis of data obtained by studying samples
inferential
Which branch of statistics helps researchers to draw inferences about the effects of sampling errors on their results described with descriptive statistics
inferential
Values obtained from a population?
parameters
Significance tests--which branch of statistics?
*inferential statistic helps researchers to decide whether the differences in descriptive statistics are reliable
Narrow bell curve?
-Small SD -Less variability
IQR 1,1,1,2,3,4,5,6,7,8,8,9,10,11,11,12
1,1,1,2 3,4,5,6 7,8,8,9 10,11,11,12 2.5--6.5--9.5 9.5-2.5=7
Three rationale of null hypothesis?
1. Observed difference was created by sampling error 2. No true difference between the two groups 3. True difference is zero
Why is range an unreliable statistic?
Based on two most extreme scores
Bell Curve: Skewed to the left
Negative Skew
SD size of tall and narrow bell curve?
Small
Use standard deviation to describe the....
variability *different groups can have same mean, but very difficult SD--variability between participants