2014 Electronic Conference on Teaching Statistics
“Bridging the Disciplines with Fun: Resources and Research”
L. Lesser & R. Reyes
The University of Texas at El Paso
D. K. Pearl
The Ohio State University
J. Weber
Georgia Perimeter College
Supported in part by Project UPLIFT:
Universal Portability of Learning Increased by Fun Teaching
NSF/EHR/DUE-1141261 (OSU), 1140690 (UTEP), 1140592 (GPC)
“bridging the disciplines…”
• Statistics is taken across disciplines
• Statistics anxiety occurs across disciplines
(e.g., in the humanities, social sciences, and in
classes for pre-service elementary school teachers)
• Fun spans disciplines
Example: “The best lyricists…are using elements of phonetics, linguistics, grammar, semantics, metrics, rhyme, rhythm,
poetics, phonology, communications, sociology, and even the psychology of verbal behavior.” (Davis, 1985, p. ix)
Background from Lesser et al.
in March 2013 J. of Statistics Education
and August 2013 JSE webinar
• fun connects to GAISE
• overview conceptualizations of fun
• 20+ modalities of fun listed
• sources of statistics fun
(e.g., https://causeweb.org/resources/fun/)
considerations for use of fun
Lesser and Pearl (2008)
• Course goal or statistical concept
associated with the item
• Where used (e.g., in class, on website)
• What to do/say/ask before using the fun
item to “set it up”
• What to do/say/ask after using the fun item
let’s illustrate using this cartoon
“It’s a Sign: A Connection between
Correlation and Slope”
(students randomly assigned to include the red part or not)
The correlation coefficient r tells us something about the
strength and linear relationship of a scatterplot of data.
By strength, we mean how tightly the points cluster
around the regression line (i.e., the line of best fit). All
else being equal, a correlation value of r = .7 (or r = -.7)
generally indicates a stronger linear relationship than a
value such as r = .3 (or r = -.3).
The direction of the relationship has to do with the sign of
r. If r > 0, we have positive correlation, which means
higher values of Y are associated with higher values of
X, and lower values of Y are associated with lower
values of X. In other words, X and Y go up and down
together. Such a scatterplot would be described best
with a line of fit that has a positive slope, and indeed this
is always the case: positive correlation happens when
the regression line slope is positive. Likewise, r < 0
means negative correlation, with X and Y moving in
opposite directions from each other, thus suggesting a
line of fit with a negative slope. Finally, a scatterplot with
no real linear trend at all (i.e., r = 0) would have a line of
fit that is horizontal, which means slope of 0. Whether
positive, negative, or zero, the sign of the correlation r is
the same as the sign of the slope of the line.
Here are lyrics to a song (sung to the tune of the familiar
folk tune “Twinkle, Twinkle Little Star” that helped you
learn the alphabet) to help you rehearse and permanently
acquire this fact in your mind:
Correlation Song (lyric © 2013 Lawrence M. Lesser)
Are points near a line, or far?
What’s the correlation, r?
If the fit supports a line,
Its slope and r would share the sign.
Twinkle, twinkle, you’re a star:
Knowing stats will take you far!
Click on this MP3 file
http://www.causeweb.org/resources/fun/db.php?id=487
so you can hear this 20-second jingle.
Now play it one more time (and sing along!).
Project UPLIFT: three urban settings
Two-Year College
University (medium-size)
University (large)
Region of U.S. Southeast
Southwest
Midwestern
Student
population
mostly Hispanic
general U.S.
demographic
Type of course Statistical literacy
Statistical literacy
Statistical literacy
Main audience General education
Pre-service teachers
Arts and humanities
Sullivan’s
Utts’
Moore & Notz’s
Fundamentals of
Statistics: Informed
Decisions Using Data
Seeing Through Statistics
Statistics: Concepts &
Controversies
Desire2Learn
Blackboard
Desire2Learn
Text
Learning
management
system
mostly Black
student-randomized experiment
(fall 2013)
(53 from comm. college; 194 from medium-sized university)
• All students asked to take pre-tests:
SATS (Survey of Attitudes Toward Statistics), SAM (Statistics Anxiety Measure)
• Half of the students randomized to have “fun
inserts” in content readings accessed via LMS
• All students take midterms/finals with
embedded multiple-choice items related to the
(12-14) content readings
• All students asked to take post SATS & SAM
POLLING QUESTION #1:
Which of these modalities of fun would be
most effective in increasing learning?
A) cartoons
B) poems
C) quotes
D) songs
% Correct with and without Song Inserts
Topic
Without song
With Song
Difference
Margin of error:
down with n
down by √n
57.3%
9.1%
61.3%
10.0%
4.0%
0.9%
Standard score
62.5%
75.0%
12.5%
Correlation & slope
60.2%
73.8%
13.6%
Equiprobability bias
40.9%
50.0%
9.1%
Multiplicity
36.1%
37.0%
0.9% (med.
Univ.)
P-value
44.4%
50.0%
5.6% (2-yr.)
Total
42.3%
50.0%
7.7%
80% CI on total difference (2.8%, 12.6%) P-value ≈ 0.04
POLLING QUESTION #2:
Why do you think
songs showed more effectiveness?
A) Time with material
B) Songs more engaging/active
C) More impact on memory
D) Other (type your idea in the question box)
Why did some fun items
appear more effective than others?
Spearman’s Rank Correlation of
“Effectiveness” with “Activeness”
from two independent raters:
0.61 and 0.66
POLLING QUESTION #3:
What dimension of the Statistics Anxiety
Measure(SAM) should be affected most by fun?
A. Anxiety
B. Attitude Towards Class
C. Fearful Behavior
D. Attitude Towards Math
E. Performance
Post Course
avg ± se
SAM and Subscale Results (lower is better)
Measure/
Subscale
Control
n = 59
Fun
n = 53
Difference
SAM total
50.61 ± 1.51
49.08 ± 1.71
1.53
Anxiety
8.32 ± 0.44
7.53 ± 0.45
0.79
Attitude Towards
Class
7.17 ± 0.26
6.40 ± 0.28
0.77
Fearful Behavior
12.86 ± 0.42
12.64 ± 0.46
0.22
Attitude Towards Math
12.46 ± 0.73
12.25 ± 0.64
0.21
Performance
9.80 ± 0.36
10.26 ± 0.39
-0.46
POLLING QUESTION #4:
Which dimension of the Student Attitudes
Toward Statistics(SATS) should be affected
most by fun?
A. Affect
B. Cognitive
C. Value
D. Difficulty
E. Interest
Post Course
avg ± se
SATS Subscale Results (higher is better)
Measure/
Subscale
Control
n = 44
Fun
n = 44
Difference
Affect
4.11 ± 0.19
4.17 ± 0.21
0.06
Cognitive
4.66 ± 0.16
4.73 ± 0.28
0.06
Value
4.84 ± 0.15
4.46 ± 0.22
-0.36
Difficulty
3.74 ± 0.15
3.75 ± 0.17
-0.01
Interest
3.69 ± 0.21
3.89 ± 0.28
0.20
Effort
6.02 ± 0.20
6.06 ± 0.17
0.04
item-randomized experiment
(spring 2014)
(≈ 1000 from larger university)
• Students take pre- and post- SATS; SAM;
and START(Statistical Thinking and Reasoning Test)
• Students see 8 content items with each item
randomized to include fun insert or not
• All students take midterm (3 items) and final
(5 items) with embedded multiple-choice
items related to the content readings
Qualitative Data
• Spring 2013 case study:
Extreme case sampling (fun-experienced
instructor) at medium-sized university, using
multiple data sources: field notes or
transcripts from 5 observations; six 1-on-1
student interviews; end-of-course ratings
and narrative comments from 21 students
• Fall 2013 data: eight 1-on-1 student
interviews from community college
Representative findings
from the qualitative data
• Contextualizing statistics problems in relatable, real-world
examples helped students either understand the concept
and/or remember the context of the content taught
• A fun/humorous approach by the instructor to teaching
statistics made the content memorable and/or tolerable for
those who experienced anxiety over taking statistics
• Almost all students were appreciative (and the exceptions
were just indifferent) of instructors who made an effort to
integrate fun/humor, real-world examples, and/or materials
during instruction to teach concepts
BACKGROUND FOR POLLING QUESTION #5:
One interviewee provided the following response to
this question: What do you think of the use of cartoon,
jokes, songs, and games by the instructor?
“I like them because I’ve been used to mathematics
classes where it’s just strict. It’s like, if you learn the
subject that’s all you can talk about in class. It’s
nice to have just …getting off topic pretty much and
kind of just letting loose for a little. It’s still related to
the subject and you don’t even realize it, it’s just
related to the subject but somehow it eases the
tension that you might think of is associated with
any math course.”
“I like them because I’ve been used to mathematics
classes where it’s just strict. It’s like, if you learn the
subject that’s all you can talk about in class. It’s nice to
have just …getting off topic pretty much and kind of just
letting loose for a little. It’s still related to the subject and
you don’t even realize it, it’s just related to the subject but
somehow it eases the tension that you might think of is
associated with any math course.”
POLLING QUESTION #5:
Which best describes how this student views in-class fun?
a) distracting
b) engaging but irrelevant
c) engaging and relevant
d) relieves anxiety
e) other (type your idea in the question box)
BACKGROUND FOR POLLING QUESTION #6:
How would you interpret the following from another
student in the case study, in response to a question on
the use of fun items, such as cartoons, songs, games,
etc.?:
“…it lightens the mood and it’s more than just
numbers being pounded into your head for an hour
and a half straight. I’ve got other professors that do
that and it’s just, frustrating after a while.”
“…it lightens the mood and it’s more than just numbers
being pounded into your head for an hour and a half
straight. I’ve got other professors that do that and it’s just
frustrating after a while.”
POLLING QUESTION #6:
What does this quote say about the student’s desire for the
classroom atmosphere?
a) Students find STEM courses too boring to promote learning
b) Students want to feel engaged regardless of ties to content
c) Students want to feel a connection to the instructor
d) Students just want to have fun
e) other (type your idea in the question box)
Key References
STATISTICS FUN:
https://causeweb.org/resources/fun/
Lesser, L. M.; Wall, A.; Carver, R.; Pearl, D. K.; Martin, N.; Kuiper, S.; Posner, M. A.; Erickson, P.; Liao, S.M.; Albert, J., & Weber, J. J. (2013). Using fun in the statistics classroom: An exploratory study of college
instructors’ hesitations and motivations. Journal of Statistics Education, 21(1), 1-33.
(also, see www.causeweb.org/webinar/jse/2013-08)
Lesser, L. & Pearl, D. (2008). Functional fun in statistics teaching: Resources, research, and
recommendations. Journal of Statistics Education, 16(3), 1-11.
AN INSPIRATION FOR THE EXPERIMENTAL DESIGN:
Garner, R. L. (2006). Humor in pedagogy: How ha-ha can lead to aha! College Teaching, 54(1), 177-180.
INSTRUMENTS (SAM and SATS):
Earp, M.A. (2007). Development and validation of the Statistics Anxiety Measure. University of Denver.
http://iase-web.org/documents/dissertations/07.Earp.Dissertation.pdf
Schau, C., Stevens, J., Dauphinee, T. L., & Del Vecchio, A. (1995). The development and validation of the
Survey of Attitudes Toward Statistics. Educational and Psychological Measurement, 55(5), 868-875.
QUESTIONS?
• Type questions you have now
in the chat window.
• If you think of questions later, please
attend the “Bridging the Disciplines”
eCOTS panel set for Friday 3:45pm EST
or email us:
dkp@stat.osu.edu, Lesser@utep.edu,
John.Weber@gpc.edu,
Rreyes9@utep.edu
THANKS FOR YOUR INTEREST!