Research Methods Final - Chapter 12

What is Pearson r?

1) Used for interval/ration data 2) Most stable measure of correlation 3) Correlation co-efficient (direction and degree of relationship)

What is percentile rank used for?

1) Used to compare student to his peers 2) Used for data on ordinal/interval scale 3) NOT often used in research

When would you use frequency?

1) Usually for demographics

What do you know if you're data set is skewed?

1) You need to use different statistical analysis 2) Can't use statistics that assume a normal curve (must use for nominal and ordinal data)

Given a mean of 15 and a median of 22 what can we say about the distribution of scores?

Negatively skewed (mean is smaller than the median)

What is percentile rank?

Percentage of scores that fall AT or BELOW a given score.

Given: Mean of: 63 Mode: 56 Median: 59 What is the skew?

Positively skewed

What are standard scores?

Reflects how many SD a student's score falls above or below the mean

What do small quartile deviations mean?

Scores are close together

What do large quartile deviations mean?

Scores are spread apart

How can you interpret the Pearson r?

Significance of "r" is based on sample size 1) Small samples sizes can have a low r and still be significant (due to more expected variation)

The z-score is an example of a:

Standard score

What is the difference between σ and SD?

σ - represents population SD SD - represents the SD of a sample

What is a z-score? (p. 330)

1) Most basic, used standard score 2) Represents SD from the mean 3) Z = (Score - Mean)/ SD

What are measures of relative position?

1) Percentile ranks 2) Standard scores (z-score, t-score)

What are measures of variability (p. 325)?

1) Range 2) Quartile deviation 3) Variance (used to calculate SD) 4) Standard Deviation (SD)

What is range best used for?

1) Rough estimate of variability of a sample (draw-back) 2) The only thing to use to describe variance in nominal data

What does a small SD indicate?

1) Scores are grouped more closely, indicating less variance in data

What is standard deviation?

1) Most stable measure of variability 2) Used with interval and ratio variables 3) Allows you to describe the data set w/mean and SD

Can standard scores be used to compare Weschler scores to College exam scores? Why?

1) No, because they use different norm groups

What are 2 tests of correlation?

1) Pearson r (interval/ratio) 2) Spearman rho (ordinal)

What are possible values for the Pearson r?

0 to +/- 1 1 = a perfect correlation

What is quartile deviation?

1) 1/2 of the difference between the upper and lower quartiles 2) (75th percentile - 25th percentile)/2 3) More stable measure of variability than the range (used when median is used) 4) Removing the outliers

What are the 4 characteristics of the normal curve?

1) 50% of scores above and below the mean 2) Mean, Median, Mode have the same value 3) Most scores are near the mean 4) ~34% of scores fall within 1SD of the mean ~14% of scores fall within the 2nd SD ~3% of scores fall within the 3rd SD

What percentile falls at the 1st SD, 2nd, and 3rd?

1) 84th %tile above the mean 2) 98th %tile above the mean 3) 99.9th %tile above the mean

Scores: 96, 84, 78, 92, and 90 Find the mean: Find the SD: Find the range: Find the median: Describe the distribution:

1) 88 2) 6.3 √[Ʃ(Score - Mean)2/n] = √[(64+16+100+4+16)/5] = √(200/5)=√40= 6.3 3) 18 4) 90 5) slight negative skew

What is variance?

1) Amount of spread among your scores 2) How far each score is from the mean 3) Ʃ(Score - Mean)2/n 4) Used to find standard deviation

What are the set-backs of using the mean?

1) Can be skewed by outliers

What is skewed distributions?

1) Data does NOT fall with-in a normal curve 2) ex. Mean, median, mode is NOT equal 3) 1 end of the data has more extreme values

What is the normal curve (p.326)?

1) Describes the normal distribution (indicating that few scores lie at the extremes)

What is the mode best used for?

1) Describing data that is bi-modal 2) Describing nominal variables

What are the draw-backs of frequency?

1) For interval/Ratio variables - data may not reoccur 2) If every value is only accounted for one time, then use alternate forms (mean, median, mode, etc)

What are types of descriptive statistics (p.322)

1) Frequencies 2) Measures of central tendency 3) Measures of variability 4) Measures of relative position 5) Measures of relationship

What is frequency? How do you find frequency?

1) How many times each value happens 2) Counting events

What are the benefits of SD?

1) Includes every score in the calculation 2) Compare one set of scores to another set of scores

What are standard scores used for?

1) Interval / Ratio scale 2) Allows comparison across tests that normed against the same population

What do measures of relative position display?

1) It's described as norm-referenced scores 2) Allows you to compare individual score to average score of the population 3) Compare 1 person's score on different measures (same subject, same time, different test)

What are measures of central tendency?

1) Mean 2) Median 3) Mode Represent the "typical" score among a larger group of scores

What is a T-score?

1) Mean score is 50 2) 1 SD is 10

David received a raw score of 56. Given a mean of 50, a median of 50, a standard deviation of 4. What is David's z score?

1.5 (56-50)/4 = 1.5

A score at the mean of the normal distribution represents a stanine of:

5

David received a raw score of 56. Given a mean of 50, a median of 50, a standard deviation of 4. Given David's information above, what is his T score?

65 T-score = (z-score * 10)+50 T-score = (1.5 * 10) + 50

In a normal distribution with a mean of 68 and a standard deviation of 4, 84% of scores fall below a score of approximately:

72

What is a stanine? (his test question)

A method of scaling test scores on a nine-point standard scale with a mean of five and a standard deviation of two

What is median best used for?

Best used when you have a wide range of scores, or an outlier

Francis wants to present the average mathematics score attained by the participants in her study. She needs to report a measure of:

Central Tendency

What are descriptive statistics?

Data analysis techniques that enable a researcher to describe many pieces of data meaningfully with numerical indices

What is range?

Difference between the highest and lowest

What is the most common graph used in research data?

Frequency polygon (p. 338)

What is mean best used for?

Interval and ratio scores 1) Most likely to be similar amongst multiple samples

The measure of central tendency used for interval and ratio data is the:

Mean

What is the most preferred measure?

Mean

Describe a negatively skewed data set?(p. 329)

Mean < Median < Mode Most scores fall to the right side

If there are extreme scores in a data set, what is the most appropriate measure of central tendency to indicate the most typical score?

Median

The measure of central tendency that represents the midpoint of the scores in a distribution is the:

Median

The median of the following distribution 1,2,4,4,5,5,5,6,6,8,9 is:

Middle number: 5

Describe a positively skewed data set?(p. 329)

Mode < Median < Mean Most scores fall to the left

What are measures of relationship? (p.332)

Tests used to find correlation/relationships between variables 1) Determine if there is a relationship 2) Determine degree of relationship

To convert a raw score to a z score, one needs the following information:

raw score, mean, and standard deviation.