Professor Rada and Professor Vogel
Remedial Speech Assessment Report for SP20
Report and data analysis prepared by Dr. Franca Ferrari-Bridgers
List your student learning outcomes as described in your syllabus. Please list ALL of the Student Learning
Outcomes that are listed in your syllabus
Gen Ed. Obj.
1.
Communicate
effectively through
reading, writing,
listening and
speaking
2. Students will
integrate knowledge
and skills in their
programs of study.
Outcome desired
Students will achieve
an increased level of
intelligibility in
speaking
Outcome desired
Students will
recognize, and
discriminate and
pronounce the vowel
and consonant
phonemes of
American English
Students will
recognize and use the
elements of syllables
and word stress,
rhythm and intonation
patterns and
incorporate them in
their conversational
speech
Students will increase
their fluency in
conversational
speech
Dr. Franca Ferrari 01/30/2016
REMEDIAL SP020
Outcome desired
Outcome desired
Page 1
Describe the assessment activity and the (student learning outcome(s) it addresses ) that occurred in your
course.
Assignment:
Students were asked to read aloud the ‘Rainbow passage’ that is an OSHA standardized test to assess articulation of
American English sounds therefore satisfying the student learning outcome “Students will recognize and discriminate
and pronounce the vowel and consonant phonemes of American English”.
Instructions: Instructors administered the test twice during the semester: the first time was during week 2 and the
second time was during week 12. Students did not have the opportunity to see the test before reading it.
List the data collection instrument (s) used for assessment.
Participants of this study were not given prior opportunity to view the reading material before the third week of class.
The students were instructed to read the rainbow passage out loud in a microphone. The students’ performances were
recorded in a soundproofed booth using the software AUDACITY. The students were not provided with verbal or visual
cues for correct pronunciation of the “Rainbow Passage”,
THE RAINBOW PASSAGE 1
When the sunlight strikes raindrops in the air, they act like a prism and form a rainbow. The rainbow is a division of
white light into many beautiful colors. These take the shape of a long round arch, with its path high above, and its two
ends apparently beyond the horizon. There is, according to legend, a boiling pot of gold at one end. People look, but
no one ever finds it. When a man looks for something beyond his reach, his friends say he is looking for the pot of gold
at the end of the rainbow.
1
For more information about the usage of the Rainbow :https://www.osha.gov/pls/oshaweb/owadisp.show_document?p_table=standards&p_id=9780
Dr. Franca Ferrari 01/30/2016
REMEDIAL SP020
Page 2
Student performance was evaluated by the instructors indicating students’ errors on instructor’s copy of the reading.
Scoring was based on the thirty-none phonemes (24 consonants and 15 vowels) of American English, as specified in the
course textbook. Sound placement and specific consonant blends were included in the tally, bringing the total number of
target sounds to 81 items. Instructors then determined the total number of items produced incorrectly by each student.
Certain sounds occurred more often in the passage than other sounds, therefore the number of possible errors per
student was 319. However, the participants were scored on the number of total errors (tokens), they made while reading
the passage.
Provide an analysis (and summary) of the assessment results that were obtained.
24 students were assessed in total. Students were part of SP020. The percentage and the total number of errors, i.e.,
number of mispronounced sounds between the first and the second reading (PRE and POST) of the Rainbow passage is
reported in the charts below.
Percentage of mispronounced sounds
30%
25%
% of errors
20%
15%
10%
5%
0%
Series1
Dr. Franca Ferrari 01/30/2016
PRE
9%
REMEDIAL SP020
POST
3%
Page 3
Total Number of Mispronounced Sounds
number of errors
600
500
400
300
200
100
0
Series1
Dr. Franca Ferrari 01/30/2016
PRE
322
REMEDIAL SP020
POST
151
Page 4
MOST COMMON TYPES OF MISPRONOUNCED SOUNDS
60
number of errors
50
40
30
20
10
0
F_θ(2)
IN_δ(11) F_m(2)
blend_F_ blend_F_
M_r(6)
nd(8)
ld(2)
F_t(10)
F_d(8)
F_s(5)
PRE
36
29
17
8
37
14
7
8
9
9
26
18
17
20
POST
3
13
3
10
21
6
1
9
4
6
9
4
16
11
F_ ʧ (2)
IN_I (8) M_I (11) M_ʊ (4) M_aɪ (6)
We performed two Paired Sample Mean t-Tests, one for the items and one for the students.
As the t-Statistics for the ITEMS analysis, t (80) = 2.94, p < 0.05, suggests the mean difference between the PRE
(M1=0.9) and the POST (M2=.03) was significant. The percentage change shows that students decreased their number of
errors by 60.5%.
As the t-Statistics for the STUDENT analysis, t (23) = 6.65, p < 0.05 , suggests the mean difference between the PRE
(M1=13.5) and the POST number of errors per student (M2=6.2) was significant. The percentage change shows that
students overall performances improved 53% between the pre and the post.
Dr. Franca Ferrari 01/30/2016
REMEDIAL SP020
Page 5
The t-Statistics together with the mean and percentage difference values indicate that students improved their
performances from the PRE to the POST. For the Item analysis, overall students decreased the number of
mispronounced sounds, suggesting that items were pronounced better in the POST. For the student analysis, the data
shows that individual students’ performance improved between the pre and the post assessment.
Describe how the assessment results that were obtained affected (or did not affect) the student learning
outcomes you identified. As part of your discussion, describe any plans you have to address the areas where
students need to improve.
The results indicate that the GEN-ED 1 learning outcomes have been generally met, with the students demonstrating a
significant reduction in the number of incorrect phoneme productions from Pre-Test to Post-Test.
Student performance indicated overall improvement in their production of the phonemes of American English.
According to Professor Vogel, the error sounds reflect the language backgrounds of the student population and the
specific phoneme errors we would expect from these Non Native Speakers. In particular, initial /r/ and /l/, final /t/, /d/,
/s/,/ld/ and /nd/ consonant blends, and vowels /I/,/aI/ and /U/ presented with many errors. The blends are typically difficult
for the Asian population. Final sound omission is typical and would be predicted for a number of language groups, as
they do not pronounce then sounds as we do in American English, or eliminate them entirely. Omission errors impact
intelligibility to a, larger degree than errors such as substitution errors.
Since the majority of our students are native Spanish and native Chinese speakers, the following areas provide problems
for them in pronunciation of words in English.
Characteristic Difficulties with English for Speakers of Chinese include:
•
•
•
•
Many dialects do not have n: for example, night becomes light
Cantonese has difficulty distinguishing l and r: for example, flied lice
There are few final consonants in Chinese – that leads to devoicing of final voiced consonants.
The final l becomes r: for example bill becomes beer
Dr. Franca Ferrari 01/30/2016
REMEDIAL SP020
Page 6
Characteristic Difficulties with English for Speakers of Spanish include:
•
•
•
•
•
•
Pronounce short i sound as a long e sound: sit and seat, sheet and ship are confused
Strong devoicing of final voiced consonants
t/voiceless th (voiceless “th” does not exist in Spanish)
When words start with /h/, the /h/ is silent
There is no /j/ (e.g. judge) sound in Spanish; speakers may substitute “y” (joke becomes “yoke”
ll pronounced as j (jump)
Action Plan:
According to Professor Rada, data analysis might go beyond determination of total number of phoneme type errors for the
entire population and each student by considering the percentage of each type of error for each student. This would
indicate a “consistency of error” (as mentioned in Rada Assessment In statute Report, 2015). This would be more
indicative of the effect of the phoneme error on the student’s overall performance.
According to Professor Vogel, reflecting on the specific errors noted in this assessment, instructors might look at the data
utilizing a phoneme process approach. For example, noting that a number of phonemes are recorded as “incorrect” in
final word position, the type of error should also be noted. Indeed, the majority of final sound errors were omissions. This
would guide the instructor is choosing remedial strategies such as increased listening/discrimination activities and
increased use of written cues. Noting which vowel sounds are produced incorrectly might indicate that a particular type of
vowel is misarticulated by a student. The instructor might concentrate on one tongue placement for that entire set of
vowels.
Discussing what pronunciation challenges the learners feel they have, and then move to the issue of syllable-final deletion
in particular.
1. We will explain why it is a problem for intelligibility, showing them examples of syllable-final deletion changing the
meaning of a word (i.e. buy vs. bite).
Dr. Franca Ferrari 01/30/2016
REMEDIAL SP020
Page 7
2. Continue to use mouth aerobics not only for muscle control, but also as a way for learners to repeatedly practice
VC and VCC sequences.
3. Do more exercises using Backward buildup: For example, the word “book” is revealed and pronounced in this way:
k --> ook --> book
Finally, in order to better understand the nature of the students’ errors we will use a four dimension rubric to evaluate the
quality of the students’ sound production. For each of the assessed sound we will assign a 0 if the sound is omitted, a 1 if
the sound is incorrectly articulated, a 2 if the sound is approaching the target sound and a 3 if the sound is correctly
articulated.
Dr. Franca Ferrari 01/30/2016
REMEDIAL SP020
Page 8
PRE
STUDENT 1
IN_ p(4)
M_p(3)
F_p(1)
blend_pr(1)
IN_b(4)
M_b(4)
IN_t(3)
M_t(5)
F_t(10)
IN_d(1)
M_d(5)
F_d(8)
Number of
incorrect
sounds
Omission =0
3
0+0
Approaching the
target sound=2
Target
sound=3
12
9
3
4
2
Incorrect
sound =1
1
1+1
0
2+2
1
blend_dr(1)
IN_k(1)
M_k(5)
Dr. Franca Ferrari 01/30/2016
REMEDIAL SP020
Page 9