Report

(For questions 1 – 5): There are many variables in Stat_Grades.sav that can be used to describe the students in
the dataset. Use the variable Section and answer the following questions:

1. Using the Variable View in SPSS (lower left), write out how Section is labeled, and note how many Section
categories are included.

The variable Section is labeled "Section of Class". In the Values category, the response modalities are
not indicated. In the Data View it is observed that this variable contains numbers. In order to know how
many categories this variable has, the corresponding frequency distribution was obtained. This variable
has three categories: 1, 2 and 3.

Section of Class

Cumulative
Frequency Percent Valid Percent Percent

Valid 1 33 31.4 31.4 31.4

2 39 37.1 37.1 68.6

3 33 31.4 31.4 100.0

Total 105 100.0 100.0

2. What type of data and level of measurement does the Section variable represent: Is the data discrete or
continuous? Is the data qualitative or quantitative? Is the data nominal, ordinal, ratio, or interval?

Section is a qualitative variable, with a nominal measurement scale. Although the response modalities
are numbers, they are only used as category identifiers. The data is discrete.

1
3. Use SPSS to output the mean, median, and mode for the Section. Include the SPSS here. Which of the three
descriptive numerical measures of central tendency (mean, median, or mode) BEST describes the variable
Section? What information does the measure tell you about the students in the dataset (Stat_Grades.sav) – be
specific?

Statistics
Section of Class

N Valid 105

Missing 0
Mode 2

The best measure described by this variable is the mode, due to its level of measurement. The
arithmetic mean and the median have no meaning, since it is a qualitative variable. Mode tells us that
section 2 is the one with the most students.

4. Which of the three descriptive numerical measures of central tendency (mean, median, or mode) LEAST (or
most poorly) describes the variable Section? Explain.

The measure of central tendency that most poorly describes the variable section is the arithmetic
mean, because it is a qualitative variable and it makes no sense to perform mathematical operations with
this variable. The values of this variable are only category identifiers, which correspond to each of the
sections.

5. Choose the most appropriate graph to represent the variable Section. (There is more than one possible
answer). Create the graph in SPSS and include it here. Be sure the graph is fully labeled. Does this graph
confirm what the mode tells you about this dataset? Explain.

2
 The bar graph confirms that the group with the highest frequency is section 2, since the bar that
corresponds to that category has greater height.

(For questions 6 – 9): There are many variables in Stat_Grades.sav that can be used to describe the students in
the dataset. Use the variable Percept (Final Course Percent) and answer the following questions:

6. What type of data and level of measurement does the Percept variable represent: Is the data discrete or
continuous? Is the data qualitative or quantitative? Is the data nominal, ordinal, ratio, or interval?

The variable is not called percept, it was called percent. The variable percent is quantitative. The data
is discrete, since the variable contains only integer values. The level of measurement of the variable is
ratio.

7. Use SPSS to output the mean, median, and mode for the Percept. Include the SPSS here. Which of the three
descriptive numerical measures of central tendency (mean, median, or mode) BEST describes the variable
Percept? What information does the measure tell you about the students in the dataset (Stat_Grades.sav)?

Statistics
Final Course Percent

N Valid 105

Missing 0
Mean 80.34
Median 82.00
Mode 78

Since the asymmetry coefficient is between -1 and -1/2, the distribution is moderately asymmetric.
Therefore, the median is the best measure of central tendency. It tells us that 50% of the students obtained
a final course percent less than or equal to 82.

3
8. Which descriptive numerical measure of central tendency is LEAST appropriate to describe the variable
Percept? Use SPSS to output ONLY that one numerical measure. Include it here. Why is that numerical
measure the least appropriate? What information does that measure tell you about the students in the
dataset (Stat_Grades.sav)?

Statistics
Final Course Percent

N Valid 105

Missing 0
Mode 78

The least appropriate measure could be the mode, since that measure is more appropriate for
qualitative variables, to know which is the most frequent category. The final course percent is a
quantitative variable, which can be best summarized by the mean or the median, depending on the
symmetry of the distribution. In this case, the final percent that is most repeated is 78.

9. Choose the most appropriate graph to represent the variable Percept. Create the graph in SPSS and include it
here. Be sure the graph is fully labeled. Discuss why you chose this graph type and what it tells you about the
students in the class.

The histogram was selected, since this is the best graph to represent quantitative variables. This graph
allows to observe the range of the variable, the symmetry of the distribution and the mode. Additionally, a
box with the mean, the standard deviation and the number of observations is included. The students in the
class obtained values of final course percent approximately between 40 and 100, the grade that is repeated
4
is slightly less than 80. The average of the students was 80.34. The distribution has a long left tail, due to
its asymmetry. In addition, the curve of the normal distribution was superimposed, so that it would be
compared with the distribution of the Final Course Percent. It is observed that this variable does not
conform to the normal distribution.

(For questions 10-13): Relationships (correlation) between variables.

10. List all five quizzes, and include for each quiz, the numerical measures of mean, mode, and standard deviation.
Explain what these measures tell you about the quizzes as compared to each other. On which quiz did
students appear to perform the best? Which quiz had the lowest performance? Which quiz had the highest
variance (variation in scores) and what does that tell you about that quiz? (100 word minimum, fully explain,
include SPSS).

Fill in this table, include the SPSS, and then explain and answer the other questions.

Mean Mode Standard Deviation

Quiz 1 7.47 10 2.481
Quiz 2 7.98 9 1.623
Quiz 3 7.98 10 2.308
Quiz 4 7.80 10 2.280
Quiz 5 7.87 10 1.765

Statistics

Quiz 1 Points Quiz 2 Points Quiz 3 Points Quiz 4 Points Quiz 5 Points

N Valid 105 105 105 105 105

Missing 0 0 0 0 0
Mean 7.47 7.98 7.98 7.80 7.87
Median 8.00 8.00 9.00 8.00 8.00
Mode 10 9 10 10 10
Std. Deviation 2.481 1.623 2.308 2.280 1.765
Skewness -.851 -.656 -1.134 -.919 -.713
Std. Error of Skewness .236 .236 .236 .236 .236
Minimum 0 3 0 2 2
Maximum 10 10 10 10 10

The students obtained the highest mean in Quiz 2 and Quiz 3. The lowest mean occurred in Quiz 1.
The most frequent grade in four of the five Quizzes was 10 points. The variability of the scores was
higher in Quiz 1. This means that the marks obtained by the students are further away from their
arithmetic mean. Distribution around the mean is in a wider range. Indeed, when detailing the minimum
and maximum value of each Quiz, it is observed that the range of values of Quiz 1 (from 0 to 10) is
greater than the ranges of Quiz 2, 4 and 5. The students seem to have the best performance in Quiz 2,
because in that evaluation they obtained the highest mean and the smallest standard deviation.

11. Looking at the relationship between each of the Five Quizzes: First, use (and include) SPSS to measure the
relationship (correlation) between each Quiz. Which two of the five Quizzes has the strongest relationship
5
 (correlation) with each other, and how do you know? Which two of the five Quizzes has the weakest
relationship with each other, and how do you know?

Correlations

Quiz 1 Points Quiz 2 Points Quiz 3 Points Quiz 4 Points Quiz 5 Points

Quiz 1 Points Pearson Correlation 1 .673** .858** .829** .504**

Sig. (2-tailed) .000 .000 .000 .000

N 105 105 105 105 105

** ** **
Quiz 2 Points Pearson Correlation .673 1 .688 .633 .700**
Sig. (2-tailed) .000 .000 .000 .000
N 105 105 105 105 105
** ** **
Quiz 3 Points Pearson Correlation .858 .688 1 .796 .493**
Sig. (2-tailed) .000 .000 .000 .000
N 105 105 105 105 105
Quiz 4 Points Pearson Correlation .829** .633** .796** 1 .445**
Sig. (2-tailed) .000 .000 .000 .000
N 105 105 105 105 105
** ** ** **
Quiz 5 Points Pearson Correlation .504 .700 .493 .445 1

Sig. (2-tailed) .000 .000 .000 .000

N 105 105 105 105 105

**. Correlation is significant at the 0.01 level (2-tailed).

The strongest correlation was presented in Quiz 1 and Quiz 3 (marked in blue, in the correlation
matrix). A correlation coefficient r = .858 was obtained, which is close to 1. The closer the correlation is
to 1, the stronger it will be. The weakest correlation occurred between Quiz 4 and Quiz 5. A correlation
coefficient r = .445 was obtained. As long as the coefficient is closer to 0, the correlation between the
variables will be weaker.

12. Use SPSS to create the scatterplot for the two Quizzes that had the strongest relationship with each other.
Paste the graph here. What does this graph show or tell you? Is it linear? Is the slope positive (going upwards)
or negative (going downwards)? Does it seem like doing well on one of the two Quizzes here relates (or
correlates) to doing well on the other Quiz?

6
 There is a positive linear relationship between the scores obtained in Quiz 1 and Quiz 3. The slope of
the line is positive which indicates that there is a positive linear relationship between these two variables.
When the Quiz 1 points increases, the Quiz 3 points also increases. When the Quiz 1 points decreases, the
Quiz 3 points also decreases. Students who scored high in Quiz 1 tend to get high scores in Quiz 3.

13. Will a low Quiz 1 grade cause a student to get a low Quiz 5 grade? Why or why not and explain? ( 50 word
minimum)

No. A low Quiz 1 grade did not cause a student to get a low Quiz 5 grade. The correlation between
variables only indicates the degree and direction of the linear relationship between two variables. It tells
us that in this case, the low grades in Quiz 1 are related to low grades in Quiz 3. The linear correlation is
not a cause-effect relationship. The occurrence of a score in Quiz 1 does not produce a score determined
in Quiz 3. In short, the correlation does not imply causality.

7
(For Questions 14-17 use the SPSS output below to determine the answers).

14. Which two variables are being investigated and label each variable as either dependent or independent.

The variables that are being investigated are Quiz 1 points (independent variable) and Final Course
Percent (dependent variable).

15. The Correlation Coefficient (or r-value) is used to measure the relationship between any two variables. In this
case, what is the r-value?

The correlation coefficient is r = .651

8
16. What does the r-value and the scatterplot tell you about the relationship between these two variables? Is the
relationship positive or negative? Is it strong, medium, or weak?

There is a moderate positive linear relationship between Quiz 1 points and Final Course Percent. The
high values of Quiz 1 points are related to high values of Final Course Percent. The low values of Quiz 1
points are related to low values of Final Course Percent.

17. Suppose a student scored a 7 on the Quiz 5. Using the scatterplot, what is a good estimate of what their Final
Course Percent?

A good estimate of Final Course Percent is about 77.

References

Gravetter, F. J., & Wallnau, L. B. (2017). Statistics for the behavioral sciences (10th ed.). Boston, MA:
Cengage Learning.

Green, S. B., & Salkind, N. J. (2014). Using SPSS for Windows and Macintosh : analyzing and
understanding data (7th ed.). Boston: Prentice Hall.