CURRICULUM-BASED MEASUREMENT NORMING FOR READING FLUENCY
AND WRITTEN EXPRESSION
FOR FRENCH IMMERSION STUDENTS IN SCHOOL DISTRICT #57
By

Sylvie St-Pierre
B. Ed. (Elem.), Universite du Quebec aMontreal, 1985
B. A., Universite de Montreal, 1982

PROJECT SUBMITTED IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF EDUCATION
m

CURRICULUM AND INSTRUCTION
THE COLLEGE OF ARTS, SOCIAL, AND HEALTH SCIENCES
© Sylvie St-Pierre, 2002
UNIVERSITY OF NORTHERN BRITISH COLUMBIA
May 2002

UNIVERSITY OF NORTHE~N
BRITISH COLU~BIA

LIBRARY

Prince Geof~e , nc

lll

ABSTRACT
Standardized tests are not always appropriate to assess French Immersion
students. In School District #57, Learning Assistance (L.A.) teachers identified the need
for an easy, inexpensive and reliable test. Curriculum-Based Measurement was a logical
choice as it is directly related to classroom materials and instruction, and it is widely used
in the English program to assess reading fluency, written expression and basic
mathematics skills.
The purpose of this project was to develop French CBM probes for reading
fluency and written expression, and to develop local norms for the French Immersion
program. The specific measures selected were Words Read Correctly, Total Words
Written and Words Spelled Correctly. Norming tables were created with the data
obtained during three norming periods. These tables will permit L. A. teachers and
classroom teachers to assess and monitor students' progress efficiently and
inexpensively.
This report explains in detail the steps taken to develop the reading fluency and
the written expression probes, the administration procedures and the scoring rules. It also
verifies the reliability and the stability of the probes over time. The various probes are
shown to be equivalent within grade.

lV

TABLE OF CONTENTS
Abstract

lll

Table of Contents

lV

List ofTables

Vl

List ofFigures

Vll

INTRODUCTION

1

METHODS
Subjects
Instruments
Development of Reading Fluency Norming Probes
Development of Written Expression Norming Probes
Procedure
Reading Fluency Probes Administration
Written Expression Probes Administration
Analysis
Ethics Approval

4
4

9
10
11
11

RESULTS
Preliminary Analysis
Demographics Analysis
Problems with Class List

12
12
12
14

Main Analysis
Descriptive Statistics
Analysis of Probe Difficulty
Analysis of Reading Fluency Probe Difficulty
Analysis of Written Expression Probe Difficulty
Reliability of the Measures
Creation of the Norming Tables and Smoothing
DISCUSSION
Issues Raised by Data
Students' Growth
Reading Fluency Probe Difficulty
Written Expression Probe Difficulty
Reliability of the Measures
· Norming Tables
Limitations

5

5
8
9

14
14
17
18
19
20
21
23
23
23
24
25
25
26
26

v
SUMMARY
Implications for Further Research
Implications for Practice

27
27
29

REFERENCES

31

Appendix A

Examples of Primary (gr. 2) and Intermediate (gr. 5)
Reading Probes (Teachers Copies)

34

Appendix B

Graphemes and Sight Words Introduced(-), Worked on(+)
and Acquired (·)from Grade One to Seven

37

Appendix C

Table of Graphemes for Grade Five and Assessment of Probes

39

AppendixD

CBM Written Expression Probes

41

Appendix E

Directions for Administration of Written Expression Probes
and Scoring procedure

45

Appendix F

Directions for Administration of Reading Probes
and Scoring procedure

49

Appendix G

Letters of Permission and Information Letter to Parents

52

Appendix H

Graph with Raw Scores and Graph with a Smoothed Curve
(Grade Six)

56

Appendix I

Descriptive Statistics of Written Expression (TWW) Results
for Three Norming Periods

58

Appendix J

Norming Table for Total Words Written (TWW) and
Words Spelled Correctly (WSC) for Grade Six

60

Vl

LIST OF TABLES
Table 1

Number of French Immersion Students (n) per School and
per Grade in SD57

5

Table 2

Number of Students (n) Being Tested During the Three
Norming Periods

12

Table 3

Percentage of Boys and Girls per Grade Participating
in the CBM Norming

12

Table 4

Number of Students (n) and Percentage for Each Reading
Fluency and Written Expression Probe

13

Table 5

Numbers of Students (n) per Grade for Each Written Expression
Probe in Fall

13

Table 6

Descriptive Statistics of Reading Fluency Results for Three
Norming Periods

14

Table 7

Descriptive Statistics of Written Expression (WSC) Results
for Three Norming Periods

16

Table 8

Reading Fluency Probe Differences Across Norming Periods
For Grade One to Seven

18

Table 9

Written Expression Probe Differences Across Norming Periods
For Grade One to Seven

19

Table 10

Pearson Correlation for Words Read Correctly Scores
Between Norming Periods

20

Table 11

Pearson Correlation for Words Spelled Correctly Scores
Between Norming Periods

21

Table 12

Norming Table for Reading Fluency- Grade Three

22

Vll

LIST OF FIGURES

Figure 1

Mean and SD by Grade and Norming Period for Reading
Fluency (WRC)

15

Figure 2

Mean and SD by Grade and Norming Period for Written
Expression (WSC)

17

1

INTRODUCTION
Learning Assistance teachers are often asked to evaluate students and establish
programs for students who are academically or behaviorally challenged. Requests for
such assessment come from teachers, parents, in-school administrators, school board
administrators and Ministry of Education officials. In order to assess a student efficiently
and accurately, Learning Assistance teachers need to use tests that have adequate
normative data, are easy to administer, inexpensive, and reliable. Often, commercial
standardized tests are used as they offer norming tables so that students can be compared
with other students of their age or grade level. However, these tests fail to demonstrate
students' progress accurately (Deno, 1985, 1992; Marston, 1989). Consequently, they are
not very practical for decision-making about a student' s instructional program.
Commercial tests have many other disadvantages: they are expensive, tiine consuming,
and not considered valid in many cases as they are not related to any specific curriculum
(Fuchs & Deno, 1994).
Curriculum-Based Measurement is a standardized measurement system that
measures basic skills in reading, spelling, written expression and mathematics. Its
assessment focuses on measurements that are observable such as counting words read
correctly in one minute and counting words spelled correctly during three minutes given
a story starter (Marston, 1989).
The advantages of CBM are numerous and important: (a) the tests are tied to the
student's curriculum (the materials in which instruction occurs) and not on a series of
problems created by commercial test developers (Deno, 1985); (b) the tests are quick to
administer and facilitate frequent administration by teachers/educators; (c) the tests can

2
have many multiple forms (Baker & Good, 1995). (d) The tests are inexpensive to
produce in terms of time and expense (Deno, 1985 ; Marston, 1989); (e) they measure
academic behaviors in the basic skills that are observable in a specific domain (Deno,
1985; Deno, Marston, Mirkin & Lowry, 1982; Marston, 1989); and (f) tests are rehttively
unobtrusive (Deno, 1985; Marston, 1989). However, it is important to remember that
CBM is one indicator and does not preclude using other specific tests to pinpoint
particular problems in academic areas (Shinn & Bamonto, 1998).
Many studies have demonstrated the reliability and validity of Curriculum-Based
Measurement as an indicator of student progress in the basic skill areas (Shinn &
Hubbard, 1992). CBM results were also compared to teachers' holistic judgement on
rating of the students' reading proficiency and were found to correlate highly (Marston,
Mirkin, Deno, 1984; Marston, 1989). Deno (1985) also found that "all ofthe curriculumbased measures were highly correlated with performance on the standardized, normreferenced tests except for the word meaning test" (p. 222).
School District #57 (SD57, 1995a) recognized all the advantages that CurriculumBased Measurement offers to teachers and specialists who make decisions about student
placement and educational programs. In order to get assistance for a student who needs
services beyond the resources available at his or her school, educators have to be able to
assess certain skills frequently to present a precise picture of the child' s performance.
CBM was adopted as it matches the problem-solving process identified by the School
Support Services (School District #57, 1996b) and the one defined by Salvia and
Y sseldyke (1991 ), i.e. , screening, program planning, pupil progress monitoring, and
program evaluation. Students of the French Immersion program were included in the

3
CBM for Mathematics (Walraven & MacMillan, 2000), but not in the earlier CBM for
Reading Fluency and Written Expression (SD57, 1996a) as English was not their
language of instruction. In its 1995 report, the district committee on the development of
local norms for Curriculum Based Measurement recommended developing such norms
for the French Immersion program (School District #57, 1995a). Some tests, such as
Bilan Qualitatif de 1' Apprentissage de la Lecture (Campeau-Filion & Gauthier, 1984) and
Test de Rendement pour Francophone (n.d.) were available but these tests were
developed for children.with French as their first language. These tests were not valid for
the French Immersion students whose first language is English. Hence, the purpose of
this project is to develop CBM norming tables for Reading Fluency and Written
Expression for the French Immersion program.

4
METHODS
This project closely follows the steps used in other projects as it is a replication of
the English CBM Reading Fluency and Written Expression (SD57, 1996b). However,
some variations in procedures are explained with more detail as the use of the French
language created the need for new scoring rules and new reading probes.
Subjects
School District #57 offers French Immersion from Kindergarten to grade 12.
French Immersion is a program for rion-Francophone children. Both parents generally
speak English at home and have at best a limited background in French. Students are
immersed in the French language from the first day of school. Most of the schooling is
done in French. English instruction starts in grade three or four, depending on the class
organization. The students have a choice of three French Immersion schools in Prince
George. Two schools offer the program from Kindergarten to grade seven and one school
from Kindergarten to grade five. The students of the latter school go then to the district
Middle school for grades six and seven. All the students continue on to secondary school
where French becomes a smaller portion of their instruction. By grade 12, students have
only one course in French.
The population for this study consisted of the entire within-district population of
the French Immersion students of School District #57. There were between 240 and 321
students tested depending on the testing period. A detailed distribution by school and
grade is shown in Table 1.

5
Table 1
Number of French Immersion Students (n) per School and per Grade in SD57
Grade

School A

School B

School C

School D

Total per

N

N

n

n

grade

1

18

29

17

-

64

2

16

14

18

-

48

3

12

19

16

-

47

4

16

6

13

-

35

5

13

II

11

-

35

6

-

12

21

8

41

7

-

16

25

10

51

Total

75

107

121

18

321

All the French Immersion students from grade one to seven were tested for both
Reading Fluency and Written Expression. Grade one students were not tested in the first
two periods, as they have not yet received sufficient instruction to develop basic skills
(Shinn, 1989).
Instruments
Development of Reading Fluency Norming Probes
Three reading probes per grade level were developed from texts used in class. All
the elementary French Immersion teachers in the district sent me a list of readers or
books that they use in their classroom during a school year. Most primary teachers use
readers from the series Je lis, j'ecris. A mots Decouverts. Contes Roses/Jaunes. Mirabelle
and Baluchon. Intermediate teachers use a variety of texts from novels, Science or Social
Studies textbooks. A selection of passages or complete stories was chosen as probe
material for each grade level. Three probes per grade were considered sufficient for this

6

project, as there were only four schools to assess and the sample size was small. Two
examples of probes, one for the primary grade and for the intermediate grade, are in
Appendix A.
The selection of probes was an important procedure as it was essential that probes
were equivalent in difficulty in each grade level (Marston & Deno, 1982). Probe
difficulty must increase as grade level increases. Tilly and Carlson (1992) recognize the
difficulty of choosing materials that were representative of each grade level expectations.
They warn "that large differences in text difficulty makes drawing valid conclusions from
student test data more difficult" (p.10). It is important to choose materials that fall within
a specific reading level.
Readability was an important step in the norming project. School District #57
used the Fry' s Readability Graph to determine the readability of their probes (SD57,
1995b). Unfortunately, this test could not be used for French texts. Thus, tables to
establish the readability had to be created. I based the criteria on a sequential table
established by the Abbotsford School District in British Columbia (n.d.). For each grade
level, their table shows the sounds introduced, worked on, and mastered during that
school year. I built tables indicating the sounds expected to be mastered in each grade
level during a school year (Appendix B). I included the sight words expected to be
mastered in grade one and grade two in their respective table. I kept passages that had
more than 80 appropriate words for the grade level in the first 100 words. There is an
example of the assessment of three probes in a detailed grade level table in Appendix C.
I used a second method to verify that the difficulty level of the probes was
Cl;ppropriately chosen for the grade. Six to seven students from various grade levels read

7

probes from a different grade level than their own. If an average reader from grade five,
assessed last year, had difficulty reading more than 20 words in the first 100 words of a
grade four probe, the probe was reconsidered for a higher level. I also showed the texts to
teachers of different grade levels and asked for feedback.
In addition to the level of difficulty, other criteria had to be considered when
choosing the probes. The stories should not be written as poems or plays, include many
unusual proper nouns, or use extensive dialogue (Shinn, 1989; Tilly & Carlson, 1992).
The length criterion was quite difficult to achieve. The guidelines suggest that texts
should be approximately 150 words in the primary grade levels (SD57, 1995b) and at
least 250 words for intermediate grade levels. Grade one to three texts were shorter as
French Immersion texts are generally shorter than comparable grade level texts for
English as a first language.
Texts chosen in this project were a combination of texts students are expected to
read in Language Arts, Science or Social Studies. There was a possibility that some texts
might have already been read in class by certain students as the materials were taken
from the classroom curricula. Fuchs and Deno ( 1994) note that degree of familiarity can
become "a source of measurement error" (p. 19) and that "conclusions about the students'
general level of proficiency could then be overly optimistic" (p. 19). Due to varying
levels of familiarity, the validity of the probes will be difficult to assess. However,
statistical tests comparing the level of difficulty among the same grade level probes will
show whether a text was possibly too familiar for

all students, or just for certain students.

Each text was retyped so that pictures would not provide clues, and so that type
style differences would be minimized. Two different copies of each text were prepared.

8

The examiner used a copy that has the number of words per line and a space to write the
results. The students read the copy with the words only.
Development of Written Expression Norming Probes
The starting sentences in the writing probes were translated from the English
language. These probes, developed by SD57 teachers, used criteria developed by Tilly
and Carlson (1992). Three out of six possible probes were chosen using two selection
criteria. The first criterion used was to have a starter that would not elicit controversial
stories. Basic vocabulary in the sentence was the second criterion chosen so that grade
one and two students would be able to understand the language. The written expression
probes are in Appendix D.
Two teachers, one teaching French Language Arts in a Secondary School and one
who taught from Kindergarten to grade five in French Immersion, met with me to
establish standardized rules for scoring the written passages of the students. We based the
rules on the Written Expression Scoring Rules established by the CBA Institute of the
University of Oregon (Baker, Collins & Goodwin, 1992). Our committee added several
rules to the list of rules defining "What is a correctly spelled word?" (p. 94-95)
established by Baker et al. For example, students had to write the acute and grave accents
correctly to get full score. Abbreviations could replace the English series of words, i.e.
NASA was acceptable. English words were rejected except for proper names such as
films, persons and cities. The committee felt that rules showing grammar should be
included in the list (Appendix E) as a reminder about looking at each word as a separate
entity regardless of the grammar.

9
Procedures
Reading Fluency Probes Administration
I translated the scoring procedure of the CBA Training Institute of the University
of Oregon (Baker, Collins & Goodwin, 1992) to determine what counted as a word read
(Appendix F). Learning Assistance teachers met for half a day to be instructed about the
administration and the scoring of the tests, and to listen to the texts read by students on a
tape recorder. Each L.A. teacher scored two readings for each of three grade levels. They
compared their scores and discussed the differences heard on the tape. Notes were taken
about possible mistakes in pronunciation, accents and liaisons (tying two words together;
for example: un ami becomes un nami). Unfortunately, there was not enough time to
listen to all the grade levels. Comments and questions were exchanged throughout the
testing period when someone was confused or uncertain about a word. After the first
norming period, one L.A. teacher was replaced by a new teacher. I reviewed the
directives and the scoring rules with her before the second testing period.
·For each grade level, students were divided into three groups: A, Band C. The
L.A. teacher assigned a letter to the students followingthe list provided by each teacher.
Each school was assigned with a different letter/ reading probe. In the fall, the L.A.
teacher from School A assigned probe A to the first student on the class list, and then
cycled through the list for the remaining students. The L.A. teacher from School B
assigned probe B first, and School C assigned probe C first. This method was used to
avoid having the biggest number of students with a certain letter if all the groups at each
grade level had an odd or even number. Students read a different probe in the winter and .

10
spring periods. For example, if a student read probe Bin October (fall), he/she .read probe
C in January (winter) and probe A in April (spring).
The first norming period was in late October. Only students in grade two to grade
seven were tested. L.A. teachers administered the reading tests in their own schools. The
L.A. teachers from School C administered the reading tests to grades 6 and 7 of School D
as there is no L. A. teacher for the French Immersion program in that school. In the last
norming period, she also administered all the reading probes in School B.
All students were recorded on tape to increase the accuracy of the scoring. L. A.
teachers were able to listen many times to a student's reading if it was not clear or too
fast. The examiners had to follow specific directions to ensure uniformity. The directions
were translated from the Directions for 1-minute Administration of Reading Passages of
Baker, Collins and Goodwin (1992). Students were asked to read to the best oftheir
abilities. They were told they would be stopped after one minute. If students hesitated
with a word for three seconds, the L.A. teachers had to say the word and mark it as
incorrect. The administration procedure and the scoring rules are given in Appendix F.
Written Expression Probes Administration
The written expression probes were administrated by each French Immersion
teacher administered the written expression probes to his or her own class. They had
specific directions (Appendix E) to follow to ascertain the uniformity of the
administration of the tests. Each school started with a different probe. All the students of
Schools A and D started with probe A. School B students started with probe Band
School C students with probe C. In January and April, students had a different starter
sentence (or probe) to write about.

11

Students were asked to continue a story starter for three minutes. They had one
minute to gather ideas. They were not allowed to ask for words in French or to get help
with spelling. Only I did the scoring to ensure uniformity in the process (Shinn, 1989). I
followed the rules defined by the committee members. At times, I found unexpected
spelling. I noted these cases to ensure similar marking in the event of future occurrences.
Analysis
All the results for both Reading Fluency and Written Expression were put into
tables per school and per grade level. I used Excel 97 to record the data and to do
statistical tests. I used the Analysis of Variance (ANOVA) for one factor to assess
probes' difficulty at each grade level. The Pearson correlation test was used to establish
the equivalence and stability coefficient for the scores over time and across probes. The
correlation value r is a measure of stability and equivalence; stability of the repeated
testing of students, and equivalence of the different probes used for testing.
Ethics Approval
The project was approved by School District #57 and the University of Northern
British Columbia Ethics Committee. In order to keep the students' results confidential,
student library numbers were used in all testings, and school names were replaced by
letters. I also sent a letter to all the parents of French Immersion students explaining the
project (Appendix G). Letters of approval are in Appendix G.

12

RESULTS
Preliminary Analysis
Demographic analysis
The number of participants ranged from 240 to 31 7 students during the three
norming periods. The results in Table 2 show that the number of students participating in
the written expression assessment is different from the number of students participating
in the reading fluency assessment in each of the norming periods. The difference in
numbers is due to absentees during testing days, and to students moving in or out of
schools. The increase in the spring is largely due to the inclusion of grade one students.
Table 2
Number of Students (n) Being Tested During the Three Norming Periods
Reading Fluency
Written Expression

Fall
n
249
244

Winter
n
240
248

Spring
n
317
300

The total number of French Immersion students registered in April was 320. The
population consisted of 47.7% boys and 52.3% girls. Table 3 shows the percentage of
boys and girls for each grade level. Most grades had a noticeable difference in the
boy/girl ratio.
Table 3
Percentage of Bovs and Girls tier Grade Participating in the CBM Norming
Grade I
Grade 2
Grade 3
Grade 4
Grade 5
Grade 6
Grade 7
Total

Number of
boys
32
18
27
16
13
22
25
153

Percentage %
50.0
37.5
57.5
45 .7
37.1
54.0
49.0
47.7

Number of
girls
32
30
20
19
22
19
26
168

Percentage %

Total students

50.0
62.5
42.5
54.3
62.9
46.0
51.0
52.3

64
48
47
35
35
41
51
321

13

The percentage of students reading each probe was as evenly distributed as
possible. In each school and for each testing period, students were evenly assigned a
different probe. Table 4 shows the number of students and the percentage of the
population for each probe and for all the testing periods.
Table 4
Number of Students (n) and Percentage for Each Reading Fluency and Written
ExQression Probe
Fall
Writing
Reading
%
%
n
n
84
33.7
73
30.0
74
30.3
83
33.3
82
33.0
97
39.8
244
100
249
100

Probe
A
B

c

Total

Winter
Reading
Writing
%
%
n
n
104
41.9
32.9
79
29.0
79
32.9
72
82
34.0
72
29.0
240
100
248
100

Spring
Reading
Writing
n
%
%
n
103
32.5
32.3
93
103
32.5
105
36.5
· Ill
35.0
31.2
90
317
100
288
100

The percentage of students using each of the three writing probes is not the same
for each probe, as the number of students varies from one school to another. Table 5
shows the number of students per probe for each grade level for the fall norming period.
In grade six, for example, there are 19 students who wrote in response to probe C and
only eight students who wrote in response to probe A in the first norming period. Only
six students used probe B in grade four.
Table 5
Numbers of Students (n) oer Grade for Each Written Exoression Probe in Fall
Grade 2
Grade 3
Grade 4
Grade 5
Grade 6
Grade 7
Total

Probe A
16
11
16
12
8
10
73

Probe B
12
18
6

11

12
15
74

Probe C
17
14
13
10
19
24
97

Total
45
43
35
33
39
49
244

14

Problems with Class List
Learning Assistance teachers (L.A. teachers) used class lists to assign probes. In
the second norming period, lists were not reviewed and new students to the school were
not included. For the last testing period, each L.A. teacher used recent lists to insure that
all students were given both tests.
Main Analysis
Descriptive Statistics
Table 6 shows various statistics for each grade level and each norming period for
Reading Fluency.
Table 6
Descriptive Statistics of Reading Fluencv Results for Three Norming Periods

Grade

Grade 1

Grade 2

Grade 3
Grade 4
Grade 5

Grade 6

Grade 7

Fall
Winter
Spring
Fall
Winter
Spring
Fall
Winter
Spring
Fall
January
Spring
Fall
Winter
Spring
Fall
Winter
Spring
Fall
Winter
Spring

-

-

Mean

22.21
29.45
42.43
55.48
57.09
67 .69
71.07
87.31
91.32
96.20
75.15
78.70
83.55
69.13
75 .64
76.54 .
77 .34
80.20
80.30

-

SD

18.58
21.01
25 .80
30.89
21.34
25.39
25.10
21.23
22.16
20.70
19.78
21.81
21.77
17.36
14.91
16.40
22 .64
22.48
22.87

-

-

Min

-

1
3
10
8
23
28
30
43
33
44
41
35
44
39
44
46
30
41
31

-

Max

75
75
140
126
117
124
-137
127
128
135
119
119
122
116
112
120
136
142
123

-

-

Skew

1.24
0.66
1.28
0.51
0.65
0.44
0.46
-0.26
-0.75
-0.19
0.36
0.05
0.13
0.59
0.15
0.47
0.09
0.51
-0.10

Kurtosis

-

-

0.87
-0.62
3.02
-0.46
0.62
-0.89
-0.09
-0.49
0.26
-0.08
-0.20
-0.66
-0.83
0.40
0.11
0.49
-0 .07
-0.05
-0.50

For the most part, the data are positively skewed across all grades and norming
periods but close to the normal distribution value of 0. The kurtosis values are a mix of

15
small positive and negative values, close to a normal distribution. In grade two, for the
fall, there is an odd value for the kurtosis and the skew: both values are very high.
The means increase from fall to winter and from winter to spring at each grade
level. There is a more noticeable increase in grade two and three from fall to winter. As
shown in Figure 1, the increase is minimal for grades four, five, six and seven between
certain testing periods. As a result, the curve on the percentile graph has some raw scores
for winter under the fall curve, or very close to it. A graph with raw scores and a graph
with a smoothed curve are in Appendix H. Smoothing was necessary and will be
discussed in more detail in a subsequent section.
Figure 1. Mean and SD by Grade and Norming Period for Reading Fluency
110
100

r

90
80
70
VJ

Q)

60

ca

50

::I

>

r
I

40
30
20

/

/"'\

\ .............. A

./

/.

~~

....

_.._

..-

10

\/

--

~
~

- __ _.,
......._

0
1 2

3

4

5

6

7

Grades

The grade four mean is greater than the grades five, six and seven means in all the
norming periods. Students read more words in grade four than in grade five, six and
seven. There is also a decrease in words read from grade five to six.

16
Table 7 shows the descriptive statistics of the results for written expression. Only
the results for the Words Spelled Correctly (WSC) are in the table. The results for Total
Words Written (TWW) are in Appendix I. They follow the same pattern as the WSC.
Table 7
Descriptive Statistics of Written Expression (WSC) Results for Three Norming Periods
Grade 1

Grade 2
Grade 3
Grade 4
Grade 5
Grade 6
Grade 7

Fall
Winter
Spring
Fall
Winter
Spring
Fall
Winter
Spring
Fall
Winter
Spring
Fall
Winter
Spring
Fall
Winter
Spring
Fall
Winter
Spring

-

-

Mean

10.40
8.47
12.78
15.91
16.28
20.91
21 .23
23 .94
30.35
31.56
30.58
36.73
. 39.31
37.49
45.43
45.49
46.47
51.82
51.94

-

SD

8.32
5.30
7.67
7.58
8.52
11.43
11.12
9.58
10.68
8.49
10.19
13.38
11.30
12.35
14.43
12.40
12.78
12.94
14.95

-

Min

1
0
1
4
2
5
6
2
5
13
7
9
19
14
18
23
26
30
27

-

Max

38
21
33
30
44
66
62
42
54
56
49
59
76
65
72
77
73
84
. 89

-

Skew

1.53
0.60"
0.79
0.31
0.97
1.42
1.19
0.09
0.00
0.28
-0.30
-0.47
0.89
0.41
0.07
0.37
0.23
0.66
0.42

Kurtosis

-

2.41
-0.64
0.01 .
-1.13
1.38
3.85
2.76
-0.03
0.21
0.91
-0.59
-0.40
1.97
-0.18
-0.91
-0.19
-1.03
-0.05
-0.42

For the most part, the data are positively skewed across all grades and norming
periods, except for two norming periods in grade five. The kurtosis values are a mix of
small positive and negative values, close to a normal distribution. There is a normal
distribution, skew of 0, in grade four for the winter testing period suggesting a normal
distribution of the curve. In grade three, the kurtosisvalues are high and they correspond
with higher positive values of skew.
The means are increasing consistently from grade one to seven as shown in Figure
2. In grades two, five and six, the mean in the fall is lower than the previous grade in the

17
spring. In all the grades, the mean increases noticeably between fall and winter but not as
much between winter and spring.

55
50
45
40
35 -

r-

~ 30

~

25

20

15
10

5
0

.... /

/

::I

...

/

_/
..Ill'

1 2

3

I
/"'

I-+~
-so-Mean -

,_/

....___

~ ---

./

,_.. .a

......-

'

4

__.....

~
'

5

'

6

'

7

Grades

Figure 2. Mean and SD by Grade and Norming Period for Written Expression (WSC)

Analysis of Probe Difficulty
It is important that probes within the same grade level offer the same challenge to
students. I used two methods to verify the probe difficulty. I used the analysis of variance
(ANOVA) test for single factor to compare the means among the probes. The ANOVA
results for reading fluency and written expression are in Tables 8 and 9. An alpha level of
.05 was chosen. In Tables 8 and 9, the letters ns mean there is no significant difference
among the three probes. The letters gg indicate that there is a significant difference
among the probes. The probes are ranked from most to least difficult.

18
Analysis of Reading Fluency Probe Difficulty
Table 8 indicates that of the 19 tests of differences for probe difficulty, there were
only two instances, spring grade two and spring grade seven, of a significant difference.
This is taken as evidence of a general lack of difference among probes within grade.
Table 8
Reading Fluencv Probe Differences Across Norming Periods for Grade One to Seven
Grade I

Grade 2

Probe
Fall

Grade 4

ns
B

c

ns
A
B

c

ns
B
A

Grade 5

Grade 6

Grade 7

Probe
Winter

Mean
Winter

Probe
Spring
ns

c

ns

c

B
A
ns

c

A
B

ns

c

B
A

25.00
28.06
35.67
50.36
55 .75
64.80
79.81
84.83
96.67
73.09
74.36
77.75
64.58
65 .77
76.69
72 .59
75.63
83.71

ns
B
A

c

ns
A

c

B
ns
B
A

c

ns
A

c

B
ns
A
B

c

ns
B

c

A

35.13
40.13
51.44
61.23
68.94
72 .62
86.75
93.64
94.00
77.67.
78.82
79.80
68.17
76.46
81.29
71.67
78.93
89.38

Mean
Spring

c

21.35
21.41
23 .89

c

39.47
55.07
73.07

A

A

Grade 3

Mean
Fall

B
Sig
A
B
ns
A
B

c

ns
A
B

c

ns
B

c

A

ns

c

B
A
sig

c

B
A

62.19
68.93
83.50
88.56
98 .94
100.39
73 .82
87.33
89.70
73.31
75.83
80.14
69.71
81.40
91.20

19
Analysis of Written Expression Probe Difficulty
I used the scores of the Words Spelled Correctly (WSC) variable to compare the
level of difficulty of the writing probes. I analyzed the results of the number of Words
Spelled Correctly, and not the Total Words Written results as both sets are highly
correlated (Marston, 1989). Analysis of one set can be transposed to the other series of
results. In Table 9, the one factor ANOVA test indicates there is a significant difference
among the probes 10 out of 18 times.
Table 9
Written Expression Probe Differences Across Norming Periods for Grade One to Seven
Probe Fall

Mean
Fall

Grade I

Grade 2

Grade 3

ns
B

c

5.75
8.71
10.25

c

11.94
17.93
21.27

A
stg
B

Grade 4

A
stg
B

Grade 5

A
ns
B

Grade 6

A
sig
B

Grade 7

c

12.83
24.00
28.06

c

25.73
31.60
34.17

c

29.58
39.63
44.25

c

40.07
46.13
56.90

A
sig
B
A

Probe
Winter

sig

c

A
B
sig

c

A
B
ns
B

c

A
ns

c

B
A
ns

c

A
B
ns
A

c

B

Mean
Winter

8.29
14.33
14.82
13.72
23.31
28.50 ·
27.27
28 .67
34.69
32.33
35.25
41.50
43 .00
44.33
51 .86
49.38
50.56
60.56

Probe
Spring
sig
B
A

Mean
Spring

c

6.80
9.67
14.50

c

12.83
16.94
17.29

c

16.53
21.06
28.73

ns
A
B
sig
A
B
ns
B

c

A
ns
A
B

27 .75
32.38
37.00

c

37.91
38.55
41.15

c

39.25
44.57
54.13

c

43.08
51.84
62.80

stg
A
B
stg
A
B

20
In the fall, the order of the probes is the same for all grades. Note that the group
that wrote in response to probe Bin the fall was given probe C in winter and probe A in
spring. The meaning of these patterns are discussed in a later section. Students who wrote
on probe A had the best results. There was a significant difference among the probes in
all the grades except for grade five.
Reliability of the Measures
I used the Pearson Correlation test to evaluate the stability of the probes over time
(from fall to spring). I compared the scores of the fall to those of winter, the scores of
winter to those of spring, and for the entire period of testing, from fall to spring.
Table 10 indicates a consistent relationship for Words Read Correctly (WRC)
between the norming periods for grades two and five. The ! value is consistent for all the
periods. In grades three and seven, the correlation is higher in the winter and spring
periods. In grades four and six, contrary to what is expected about the progression of the
· reading skills, there is a lack of decrease in the fall-spring (6 month) period. The highest
correlation value is in grade five.
Table 10
Pearson Correlation for Words Read Correctlv Scores Between Norming Periods
Grade

Grade 2
Grade 3
Grade 4
Grade 5
Grade 6
Grade 7

r
Fall- Winter
(3 month)
.77
.79
.74
.87
.76
.68

* Grade one was tested only m Sprmg

r
Winter- Spring
(3 month)
.78
.80
.73
.87
.69
.81

r
Fall- Spring
(6 month)
.77
.70
.80
.86
.79
.74

21
Table 11 indicates a higher correlation between the Words Spelled Correctly
(WSC) scores in the winter-spring period for all the grades. It decreases in the fall-spring
period for most grade levels as the period covered is larger, thus more prone to
fluctuation in the students' writing abilities or motivation. The r value increases in grade
five only, going from .52 to .60. The r values for writing are lower than the r values for
reading, thus less stable.
Table 11
Pearson Correlation for Words SQelled Correctlv Scores Between Norming Periods
Grade
r
r
r
Fall- Winter
Winter- Spring
Fall- Spring
(3 months)
(3 months)
(6 months)
.57
.71
.48
Grade 2
Grade 3
.73
.75
.72
Grade 4
.42
.45
.30
Grade 5
.32
.52
.60
Grade 6
.66
.77
.65
Grade 7
.49
.84
.56
* Grade one was tested only m Spnng
Creation of Norming Tables and Smoothing
In order to build norming tables, all the students' results were ranked from 1 to 99
percentile. Raw scores and percentiles of each testing period were put on one graph for
each grade level. The resulting graphs show the students' improvement in the skills tested
through the testing periods. Unfortunately, some results were less than to the results of
the previous testing period. As a result, some curves were below the curve of the next
norming period. Hence, the curves needed to be smoothed. I hand-smoothed most curves
in the three categories reported on, Words Read Correctly (WRC), Total Words Written
(TWW) and Words Spelled Correctly (WSC).
I generally lowered the scores in the fall and winter, instead of increasing them,
so that cut-off scores are not artificially inflated. I changed the scores as minimally as

22
possible, to just below the score of the later testing period. I had to make so many
changes that I did not try to make the curves as smooth as possible. I did not modify any
graph for grade one as students read and wrote only in the spring.
I created the norming tables using the data of the modified graphs. Some
percentile data were available directly from the results. Some other data needed to be
pulled from the curve. Table 12 is an example of a norming table for reading fluency for
the grade three students. The two set of scores for written expression, i.e., Total Words
Written (TWW) and Words Spelled Correctly (WSC) are in .the same norming table for
each grade. An example is in Appendix J.
Table 12
Norming Table for Reading Fluency- Grade Three
Percentile

99
95
90
85
80
75
70
65
60
55
50
45
40
35
30
25
20
15
10
5
1

Fall
WRC

112
88
80
76
71
68
66
64
63
61
57
56
52
47
45
42
39
35
31
27
24 .

Words Read Correctly- Grade Three
Spring
Winter
WRC
WRC

121
109
101
94
88
83
78
75
70
65
60
58
55
53
51
50
48
40
38
34
28

129
115
102
96
90
88
84
79
76
72

70
68
64
61
57
54
51
43
38
36
32

Description

Well above average
Above Average

Average

Below Average

Well Below Average

23

DISCUSSION
Many statistical tests are required to develop reliable assessment tools. If the tools
are reliable, the norming tables emerging from the results can be considered reliable as
well. Consequently, it is important to examine the results of the statistical tests performed
in this project. The sensitivity of the probes to students' growth, the difference between
the probes and, finally, the reliability of the probes will be discussed.
Issues Raised by Data
Students' Growth
Overall the curve of the means for reading fluency (Figure 1) shows a normal
upward curve that plateaus in grades six and seven. The curve for CBM reading fluency
in English also shows an increase from grade one to seven, with a slower growth rate at
the upper intermediate levels (SD57, 1996a). The two curves were judged to be similar.
In grades two and three, the curve shows a steady improvement from one period
to the other. In these grades, much emphasis is on reading skills instruction, with the
result that students demonstrate considerable reading growth in a short time period.
Grades six and seven students show less growth in their reading skills, as indicated by the
plateau shown on the curve. This corresponds to a shift in reading instruction focus. In
the upper intermediate grades, the emphasis is on reading as a way to obtain information
through the application of skills taught in earlier grades.
There is a constant increase from grade one to seven with the written expression.
This was not observed with the English written expression (SD57, 1996a). There is a
second effect that is apparent. Students improve between each period, but more
noticeably between fall and winter. The means in the fall for grades two, five and six are

24
slightly inferior to the mean of the previous grade level for the spring period. This
phenomenon is called the Summer Effect. Students don't use their writing skills in the
summer as much and their spelling is not as sharp as in winter or spring. The saw tooth
effect is similar to the one observed with the English CBM.
Reading Fluency Probe Difficulty
Comparable assessment instruments are essential when they are used to compare
students' abilities. It is then important to build instruments that are comparable in
difficulty over extended periods of time, as they offer "several instructionally related
advantages over measuring student performance on ever-changing or increasingly more
difficult samples of material" (Fuchs & Deno, 1994, p. 20). The results obtained from the
methods used to verify the level of difficulty of the probes have an impact on the
effectiveness and the reliability of the norming tables created in this project. The results
show no significant differences among the reading fluency probes for all the grade levels,
indicating that probes are of equal difficulty. Although the English reading fluency
probes within grade were judged to be of equal difficulty, greater variation among theses
probes was apparent.
Using texts from the classrooms' curricula brings a risk of having some students
being familiar with a probe. The results could be affected by this situation and the
norming scores could be higher than if all students would read a new text, especially if
the sample is small. However, it is part of the CBM characteristics to use texts studied in
class. This situation happened during the course of this project. In grade two, three
students had read one of the texts. However, their scores were average and did not stand
out from the others. It did not affect the results.

25
Written Expression Probe Difficulty
The selection of probes is also important for the written expression norming. The
results were greatly affected by the group of students using a specific probe. In the first
norming period, probe A has the highest mean in all the grades. These students, with the
exception of grade four and five in winter and grade four in spring, always had the most
words correctly spelled. It can be explained by the fact that they are all from the same
school and that they participated in a program to enhance written expression. They had
used the program Write On (Hart, 2001) since the beginning of the school year. The
students who had the least correctly spelled words except for grade four (winter and
spring) and grade seven (winter) are from the same school. The probes' order changes but
the groups using them always stay in the same position. Even though the ANOV A test
suggests a difference in difficulty between the probes in 10 instances out of 19, the
difference can be attributed to the strength of the students. It is evident that a school
· effect has influenced the scores.
Reliability of the Measures
The! values demonstrate that the reading fluency probe-s were stable over time
and across probes. The range ofr values is wider for the French Immersion CBM (.68
to.87) than for the! values for the English CBM (.77 to .89) reported in the school district
technical report (SD57, 1996a). The median in English is .85 and the median in French is
.78. Values from the Pearson test demonstrate that the reading probes are stable over
time. I regard these two values as equal.
The correlation between the written expression probes is not as satisfactory as the
reading probes. The! values for the Words Spelled Correctly vary greatly from one

26
period to the other for all the grades, except for grade three. The range of r values is from
.30 to .77. Similar data for the English CBM are not available for comparison. However,
the correlation coefficients for Total Words Written are available; they range from .48 to
.69 (SD57, 1996a). The median value in English is .62 and the median in French is .58;
they are both similar. In both instances, the correlation coefficients are lower than the
values for reading.
The decrease in stability for the written expression probes can be explained by the
fact that reading and writing are different cognitive activities. Reading a text for one
minute does not require any creativity or inspiration as writing does. Therefore, students'
performance in written expression can be affected by their mood or motivation on any
given day. Hence, more uneven performances are to be expected.
Norming Tables
We were successful in our attempt to create local norms for the French Immersion
program. The regularity of the smoothed curves across testing times and the similarity to
the English norming curves are a demonstration of this.
Limitations
The population of the students registered in French Immersion was quite small
with a maximum of 321 students. In order to have more students in the study, the
researcher would have had to include French Immersion students from another school
district. This option would have contradicted the purpose of establishing local school
district norming tables and created many logistical problems.
The number of students varied for each testing period due to student absentees.
The limited time allotted to French Immersion Learning Assistance teachers in each
school did not permit the testing of students who were absent for a long period.

27

SUMMARY
The objectives of this project were to develop efficient and inexpensive tools for
Learning Assistance teachers to use with French Immersion students, and to create
norming tables for reading fluency and written expression. The project was based on
standard methods used in the School District #57 norming study (1996a). Three reading
fluency probes for grades one to seven were developed. The ANOVA test shows no
significant difference in difficulty between the probes. The written expression probes
show significant difference in difficulty for many probes but the grouping of the students
had an effect on the results as one group was stronger in written expression skills. Based
on all the preceding analysis, both sets of probes, reading fluency and written expression,
are considered reliable testing instruments.
The norming tables created in this project are a tremendous asset for the French
Immersion program. They provide logical and usable norms for reading fluency and
written expression.
All the results show similar reliability to the previous English study (1996a).

Implications for Further Research
French Immersion students start receiving formal English Instruction in grade
three or four depending on the composition of their classes (split or one grade only). Even
though they can read in English at that point, they do not spell in English as well as their
peers who have had English instruction since Kindergarten. Anecdotal discussions
suggest that by grade seven, French Immersion students have caught up with their
counterparts from the English program. It would be interesting to compare the French
Immersion students' CBM scores for reading fluency and written expression to ·CBM

28

scores in their first language after English instruction has begun. One might suppose that
they would write more in their first language as they are more comfortable in English.
One might also expect no significant difference between their reading fluency scores as
they have read and been read to in English from a very young age.
While much research has demonstrated the "validity of CBM reading as a
measure of general English reading proficiency, including comprehension" (Marston &
Deno, 1982; Shinn & Good, 1992), there is no general consensus about second
languages. A study by Baker and Good (1995) provides "initial support for the validity of
CBM reading as a measure of English reading proficiency, including reading
comprehension, for bilingual students."(p. 572). However, Bertin (1988) demonstrates in
her study that knowing the linguistic code and the meaning of each word individually are
not sufficient to comprehend the text as a whole. As a fifteen year experienced educator, I
support Bertin' s view. Students may be able to decode extremely well but may not know
the meaning of the words or the idea expressed. Developing a French CBM test for
comprehension like a cloze-test and comparing the scores with CBM scores for reading
fluency could enhance our knowledge about this topic.
CBM has proven reliable for assessing the basic skills of students in French
Immersion. If these scores compared to the achievement grades letters French Immersion
students receive at school, would they be highly correlated? Do the CBM scores give a
realistic picture of the academic performance of the students? It would be interesting to
compare letter grades of academic subjects, such as Language Arts, Social Studies,
Mathematics and Science to CBM scores for reading fluency and Mathematics. That

29

research was done for the English students and indicated positive correlation between
CBM scores and grades (Fewster, 2000).
Implications for Practice
Creating norming tables for French Immersion is certainly an asset for French
Immersion L.A. teachers and classroom teachers. The norming tables are easy to use and
results give a good idea where a student' s basic skills are compared to others. Teachers
often need to evaluate the progress of a grade one or grade two student and help parents
make a decision for his or her future placement. It is in the early primary grades that
schools, parents and teachers prefer to decide if French Immersion is the right program
for their child. CBM testing is quick to conduct and teachers themselves can administer it
frequently.
CBM is also a useful tool for quickly assessing quickly a large group of students.
In the fall, teachers often do not know all their students' abilities and can not ask for
support from their L. A. teacher at that point. The scores generated by CBM allow them
to quickly evaluate the students and give the necessary support early in the year.
My data are available for research study comparing boys to girls' scores.
Knowing if one of the genders has much lower scores in reading fluency or written
expression could help teachers evaluate and adapt their teaching strategies to the gender
composition of their class.
The norming tables created in this project were designed to assess students of
School District #57. Exporting the norming tables to other districts would have to be
done with caution. Other school district French Immersion programs are in a different
context which may affect the learning progression of their students. For example, school

30
districts might have only one French Immersion school, a big tum over of teachers every
year, many trilingual students, an advisor, full-time L.A. teachers, and so on. The
norming tables might not be reliable in these different contexts. However, I am available ·
to help any school district who may want to develop CBM probes and norming tables for
their French Immersion students.

31

REFERENCES
Abbotsford School District (n.d.). Sequential Grapheme Table. Abbotsford, B.C.: Author.
Baker, S, Collins, V. & Goodwin, M. (1992). Administration and scoring of CurriculumBased Measurement. CBA Training Institute, University of Oregon, p. 88-89
Baker, S. K., & Good, R. (1995). Curriculum-Based Measurement of English reading
with bilingual Hispanic students: A validation study with second-grade students.
School Psychology Review, 24 (4), 561-578.
Bertin, C. (1988). Le role des strategies de lecture dans la comprehension des textes en
langue etrangere (The role of reading strategies in foreign language text
comprehension), Canadian Modem Language Review, 44 (3), 527-535.
Campeau-Filion, F., & Gauthier, G. (1984). Bilan Qualitatif de 1' Apprentissage de la
Lecture CBOAL) [Qualitative status of reading acquisition] (2"d ed.). Manuel de
l'examinateur [Administratormanual]. Presses de l'Universite du Quebec.
Deno, S. L. (1985). Curriculum-Based Measurement: The emerging alternative.
Exceptional Children, 52 (3), 219-232.
Deno, S.L. (1992). The nature and development of curriculum-based measurement.
Preventing School Failure, 36 (2), 5-16.
Deno, S, Marston, D., Mirkin, P., and Lowry, L. (1982). The use of standard tasks to
measure achievement in reading, spelling, and written expression: A normative
and developmental study. (Research Report No. 87). Minneapolis: University of
Minnesota, Institute for research on Learning Disabilities.

32

Fewster, S.A. (2000). School-based evidence for the validity of curriculum-based
measurement norms in School District #57. Master' s thesis, University of
Northern British Columbia, Prince George, British Columbia, Canada.
Fuchs, L., & Deno, S. (1994). Must instructionally useful performance assessment be
based in the curriculum? Exceptional Children, 61 (1), 15-24.
Hart, N. (2001). Write On. Prince George, B.C.: School District #57 (Prince George).
Marston, D.B. (1989). A Curriculum-Based Measurement approach to assessing
academic performance: What it is and why do it. In M.R. Shinn (Ed.),
Curriculum-based measurement: Assessing special children (pp. 18-78). New
York: Guilford Press.
Marston, D., & Deno, S. L. (1982). Implementation of direct and repeated measurement
in the school setting. Minnesota University. Minneapolis Institute for Research on
Learning Disabilities (ERIC Document Reproduction Service No. ED 226048)
Marston, D., Mirkin, P., & Deno, S. (1984). Curriculum-Based Measurement: An
alternative to traditional screening, referral, and identification. The Journal of
Special Education, 18 (2), 110-117.
Microsoft (1997). Excel 97 [Computer software] .
Salvia, J., & Ysseldyke, J. E. (1991). Assessment (5th ed.). Boston: Houghton Mifflin.
School District #57. (1995a). Report ofthe committee on the development of local norms
for Curriculum Based Measurement. School District #57. Prince George, BC:
School District #57.
School District #57. (1995b). Curriculum Based Measurement. School District #57.
Prince George, BC: School District #57.

33
School District #57. (1996a). Guidebook for the use of Curriculum Based Measurement
in School District #57. Prince George, BC: School District #57.
School District #57. (1996b). School Support Services: Practices, organization, and
principles. Prince George, BC: School District #57.
Shinn, M. R. (1998). Identifying and defining academic problems: CBM screening and
eligibility procedures. In M. R. Shinn (Ed.), Curriculum-Based Measurement:
Assessing special children (pp. 90-129). New York: Guilford Press.
Shinn, M. R., & Bamonto, S. (1998). Advanced applications of Curriculum-Based
Measurement: "Big ideas" and avoiding confusion. In M. R. Shinn (Ed.),
Curriculum-Based Measurement: Assessing special children (pp. 1-29). New
York: Guilford Press.
Shinn, M . R. & Hubbard, D. D. (1992). Curriculum-Based Measurement and problemsolving assessment: Basic procedures and outcomes. Focus on exceptional
Children, 24 (5), 1-20.
Test de Rendement pour Francophone (Performance Test for Francophone). (n.d.).
Tilly III, W. D. & Carlson, S. (1992). Creating measurement materials. CBA Training
Institute, University of Oregon.
Walraven, G. & MacMillan, P. (2000). Curriculum Based Measurement norming for
math (calculation) in School District 57. University ofNorthem British Columbia,
Prince George, British Columbia, Canada.

34

Appendix A
Examples of Primary (gr. 2) and Intermediate (gr. 5) Reading Probes
(Teacher's Copies)

35

Nom:
Niveau:
Date:

Dans rna maison.
Dans le salon, on se repose, on lit et on parle.
C'est la qu'on reyoit la visite.
C'est la aussi qu'on regarde la television.
Voici la cuisine.
Papa et maman travaillent beaucoup ici.
Souvent, la table est dans la cuisine.
Alors, la famille mange dans la cuisine.
L 'heure du repas est un moment bien agreable.
Dans la chambre a coucher, il y a unlit pour dormir.
On se repose et on s' amuse aussi dans sa chambre.
Mon ami a des lits superposes.
J e suis bien dans mon lit quand il fait froid.
Les lits ont des couvertures chaudes et un bon oreiller.
J e fais souvent de beaux reves.
Souvent, je me cache sous le lit.
Je saute aussi dessus.
C'est agreable de lire au lit.
J'aime bien parfois manger dans mon lit.
Ce qui est important, c'est que je peux y dormir.

Nombre de mots lus:
Nombre d'erreurs:
Nombre de mots Ius correctement:

2-C

3
14
20
27
30
36
43
50
58
70
80
86
96
106
112
119
123
129
136
146

36

Nom:
Niveau:
Date:

Un loup-garou comme mari?
Il etait une fois, au Nord du Grand Lac des Esclaves,
dans le Territoire du Nord-Ouest, une jeune femme
inuit qui habitait avec ses deux freres. Elle n'etait pas
encore mariee. Ses deux freres s'inquietaient pour son avenir.
Or, un jour un tres bel homme vint leur rendre visite.
Elle rut vite impressionnee par la gentillesse de cet inconnu.
Ce jeune homme se montra tres habile et tres responsable.
Les deux freres l'aimerent bien. Ils crurent qu'il serait un
bon mari pour leur petite soeur. Quelques semaines apres,
le jeune visiteur demanda a cette jolie dame de 1'epouser.
Elle aimait beaucoup et accepta dele marier.
Heureux, les epoux emmenagerent dans leur maison.
La nuit de noces, la mariee rut soudainement reveillee
par des bruits. Elle entendit des hurlements. Lajeune epouse
voulut que son mari aille voir ce qui se passait dehors.
A sa grande surprise, son mari n' etait pas Ia. Elle dec ida
done d'aller voir d'ou venaient ces bruits. Ellene vit rien.
Elle retouma se coucher. Quelques heures plus tard,
elle entendit encore des bruits. Mais, cette fois-ci,
ils venaient de 1'etable. Les animaux s' agiterent. Pour se
proteger, elle apporta une fourche. Un animal etait la . . .
c'etait un loup.

Nombre de mots Ius:
Nombre d'erreurs:
Nombre de mots Ius correctement:

5- c

5
16
25
35
44
55
65
75
85
94
104
112
119
128
138
149
160
171
179
188
197
206
209

37

Appendix B
Graphemes and Sight Words Introduced(-), Worked on(+) and Acquired(-)
from Grade One to Seven

38

sounds
Letters sounds
Simple syllable Ex: ba, mo
Inverse syllable: Vowel- consonnant II Cons.- vowel- cons.
Simple sound
Ex: on, ou, en, in, un, an, er, et, ai,
ei, au, oi, ch, qu, ui, eu
Silent "h"
Sound "o" open and close
Complex sounds: ex: ph, eau, ain,
em
Syllables with 2 consonants
Ex: br, fl, gr, ...
M before p and b
Ex: campagne, chambre
Complex sounds: eu, oeu. Ex: oeuf
Complex sounds Ex: noeud
S=Z Ex: fraise
Ent - ending in verbs Ex: aiment
Ent- as a sound Ex: lentement
"tion"
" ille"
"ail", "aille"
"eil", "eille"
"ouille", "ouil"
"oeil", "euille", "ueille", "eil"
Sight words 1

A, a, aller, allons, arrive, au; aussi, autre,
avec, beau, belle, bien, bon, bonjour, ce,
bonne, ces, cette, comme, content, deux,
contente, dans , de, des, dit, du, elle, elles,
en, et, faire, fait, gros, il, il y a, ils, j'ai, je,
je suis, jouer, Ia, le, leur, leurs, lui, rna,
mais, manger, marche, mes, moi, mon, ne
pas, notre, nous, on, ou, par petit, petite,
plus, pour, pres, qu ', que, qui, sa, sans, si,
son, sur, ta, te, toi, tous, tout, toute, tn!s, tu,
un, une va, cas, vous, votre, viends, venez,
vite, voir, vois, voit, veut, veux, y

Sight words 2

Grade
1

+

Grade

2

-

+

-

+

-

+
+

-

Grade
3

-

+

-

+

-

+

-

-

-

+
+
+

-

-

-

+

Aux, achete, appelle, apportons, autour, autre, bientot,
cette, ces, chante, chez, chien, combien, depuis,
derriere, ecris, en, enfants, faites, froid, il y a, jamais,
jouons, leur, lis, monsieur, neige, nos, on, par, parce
que, peur, plus, pourquoi, pris, que!, qu 'est-ce que
c' est, sait, sans, sien, soi, tire, toujours, toute, travail,
tres, va-t-en, vieux, voici, vos

-

+

Grade
4

+
+
+
+
+
+
+
+
+
+

-

Grade
5

+
+
+
+
+
+
+

Grade

Grade
7

+
+

+
+

6

39

Appendix C
Table of Graphemes for Grade Five and Assessment of Probes

40

Sounds
Complex sounds
Ex: ph, eau, ain, ein
Syllables with 2
consonants
Ex: br, fl, gr, ...
M before p and b
Ex: campagne, chambre
Complex sounds: eu,
oeu. EX: boeuf,
Complex sounds: eu,
oeu. Ex: noeud,
"s" = "z" ex: fraise
"ent" - ending in verbs
Ex: ils marchent
"ent" - as a sound
Ex: lentement
"tion"
"ille"
"ail", "aille"
"eil", "eille"
"ouille", "ouil"
·"euil", "euille", "ueille"
"veil", "oeil"
Others

(-) introduced
(+)to work on
(.) acquired

Probe A
Mon
anniversaire

Probe B
Mouvements
de l'ocean

Probe C
Un loup-garou
commrne mari?

8

6

5

7

1
1

+
+

2

3

1

1

+
+
+
+
+

8

2
3

2
1
1

1

magnifiques

4
d' est, en est,
. equateur,
superficiels

Words with more than
3 syllables

4
magnifiques,
merveilleuse,
marguerites,
anniversaire (2)

Hard words (others and
words with> 3 syl. not
included in others)
Easy words

5

4

95

96

1

superficiels

7
s'inquietaient,
impressionnee,
esclaves, inuit,
homrne, or,
femme
2
Territoire,
s' inquietaient,
impressionnee,
gentillesse,
responsable
10

90

41

Appendix D
CBM Written Expression Probes

42

Numero de l'eleve:
Niveau:
Date:

A

Ecris une histoire qui commence par ...
Hier, un singe est entre dans l'ecole en passant par la fenetre et ...

TME:
TMBE:

43

B

Numero de 1'eleve:
Niveau:
Date:
Ecris une histoire qui commence par ...
Je jouais dehors lorsque tout a coup des extra-terrestres ...

TME:
TMBE:

44

c

Numero de l'eleve:
Niveau:
Date:
Ecris une histoire qui commence par ...
J'ai trouve un crayon magique et ...

TME:

TMBE:

45

Appendix E
Directions for Administration of Written Expression Probes
and Scoring Procedure

46

Directives pour !'administration du test CBM d'ecriture

(Directions for Administration of Written Expression Probes)

Materiel: test, chronometre, les directives
1-

Assurez-vous que les eleves aient un crayon bien aiguise et une gomme a effacer.

2-

Assurez-vous qu' ils connaissent leur numero d'eleve.

3-

Donnez les feuilles de test aux eleves et demandez-leur d'ecrire leur numero
d'eleve et de toumer la feuille a l'envers lorsqu'ils ont fini .

4-

Dites ces directives (n'ajoutez pas d'autres directives, s.v.p.)
Vous allez ecrire une histoire. Premierement,je vais vous lire une phrase qui
est un debut d'histoire. Vous allez ensuite continuer l'histoire, (ecrire ce qui
arrive apres ce debut d;histoire). Je vais vous donner 1 minute pour penser a
l'histoire et ensuite vous allez ecrire pendant 3 minutes. II ne faut pas oublier
qu'il faut ecrire le mieux possible. Si vous ne savez pas comment ecrire un
mot, ecrivez-le du mieux possible, comme vous pensez. Vous ne pouvez pas
me demander des mots en f ~
(suivez bien cette regie- cela va ralentir leur
ecriture s'ils vous posent des questions- 3 minutes d'ecriture sans interruption) .
Faites du mieux que vous pouvez. Avez-vous des questions? (pause)
Voici le debut de l'histoire: ...

5-

Apres avoir lu le debut de l'histoire, laissez les eleves reflechir pendant 1 minute.
Verifiez qu'aucun eleve ne commence a ecrire.
Apres 30 secondes, dites: vous devriez etre en train de reflechir a ... (phrase de
l'histoire)

6-

Apres que la minute soit terminee, dites: Commencez a ecrire maintenant.
Commencez a compter la periode de 3 minutes.

7-

Encouragez les eleves a se concentrer sur le travail a accomplir.

8-

Apres 90 secondes, dites: vous devriez etre en train d'ecrire au sujet de (sur)
... (debut de 1' histoire)

9-

Apres 3 minutes, dites: Arretez, deposez vos crayons. Vous avez bien travaille.
Me rei.

10-

Ramassez les feuilles. Asstirez-vous qu' ils ont ecrit leur numero d'eleve

47

C.B.M. - Ecriture - Regles de correction
(Scoring Procedure)

Le test verifie deux habiletes en ecriture: l' aisance a ecrire et l' epellation. Pour
que les resultats soient comparables entre les ecoles, il est important que les
correcteurs/trices appliquent les memes regles pour compter le nombre total de mots
ecrits et le nombre total de mots bien epeles. En cas de doute, il faudra contacter la
coordinatrice du projet.

1-

Comment compter les mots:

a) L ' apostrophe: une lettre suivie d' une apostrophe sera compte pour un mot carla lettre
remplace un mot.
Exemples: L' ecureuil - 2 mots (le ecureuil)
J'ai- 2 mots Ge ai)
Ce n'est pas- 4 mots (ce ne est pas)
b) Le trait d ' union: si le trait d'union unit 2 mots qui existent separement, on comptera 2
mots pour ce groupe. Par contre, si un des mots n'existe pas separement, on comptera
1 mot pour ce groupe de mots.
Exemples: Porte-feuille- 2 mots
Peut-etre- 2 mots
Est-ce que- 3 mots
Pre-test- 1 mot
c) Les mots anglais: chaque mot sera compte.
d) Les mots inventes: les mots inventes seront comptes.
e) Les abreviations: les abreviations seront comptes pour un rriot
Exemples: BBQ - 1 mot
T.V. ou tv- 1 mot
CD -1 mot
f) Les nombres: les nombres doivent etre ecrits avec des lettres.
Exemples: Les 2 chaises - 2 mots
J'ai 8 ans- 3 mots
J'ai vingt-cinq ans- 5 mots
2-

Comment compter les mots bien epelles:

On ne doit pas tenir compte des differentes regles de gramrnaire de la langue fran<;aise .
a)

L ' accord des verbes au sujet: si le verbe existe avec l' epellation du texte, il sera
compte comrne "bon". Il ne doit pas s'accorder avec le sujet.
Exemple:
Ils mange - 2 mots bien epeles
Nous avons bus - 3 mots bien epeles

48

b)

L' accord des verbes aux temps et modes: si le verbe existe avec l' epellation du
texte, il sera compte comme "bon". Une mauvaise utilisation du temps ou du
mode de verbe ne doit pas etre penalise.
Exemples:
Ils ont mange - 3 mots bien epeles
Vous pense - 2 mots bien epeles
J' ai mettre - 3 mots bien epeles

c)

L'accord des adjectifs et des noms: les noms et adjectifs ne seront pas penalises
s'ils ne sont pas accordes aux genres (feminin/masculin) et au nombre
(singulier/pluriel)
Exemples:
Les beau fille- 3 mots bien epeles
Le petite chaises vert - 4 mots bien epeles

d)

Les determinants: le choix errone d'un determinant (article, adjectifpossessif) ne
sera pas penalise si le mot est bien epele.
Exemples:
Ma chapeau- 2 mots bien epeles
La ecole - 2 mots bien epeles

e)

Les homophones: le choix errone d'un homophones ne sera pas penalise si le mot
utilise est bien epele.
Exemples:
Les chiens on mange 1'heure os. - 6 mots bien epeles
Mait yeux sons bleus - 3 mots bien epeles

f)

Les accents: les accents sur les voyelles doivent etre precis (bonne direction).
Exemples:
Tres fache- 1 mot bien epele
Le gateau - 1 mot bien epele .

g)

Les noms propres de personnes ou d'animaux: si l'orthographe des noms propres
de personnes ou d' animaux varie dans le texte, 1' orthographe du mot le plus
utilise sera pris en consideration.
Exemples:
Sabrina, Sabrinae, Sabrinae - 2 mots bien epeles

h)

La majuscule au debut d'une phrase: on ne comptera une faute si le premier mot
de la phrase n'a pas de majuscule.

i)

Les lettres inversees: si la lettre inversee produit un mot bien orthographie, ce mot
sera accepte. Mais si la lettre produit un inot qui n' existe pas ou qui n' est pas bien
ecrit, il ne sera pas considere comme mot bien epele.
Exemples:
Doule au lieu de boule - 0 mot bien epele
Bon au lieu de don - 1 mot bien epele

j)

Les abreviations: les abreviations sans points entre les lettres seront considerees
comme bien orthographies.
Exemple:
La tv - 2 mots bien epeles
La NASA - 2 mots bien epeles

49

Appendix F
Directions for Administration of Reading Probes
and Scoring Procedure

50
Directives pour I' administration du test CBM de lecture
(Directions for Administration of Reading Probes)
Materiel:
Copie de leleve (probe A, B ou C)
Copie du professeur (avec le nombre de mots par ligne)
Un chronometre
Un enregistreuse a cassette
1-

Placer Ia copie de I' eleve devant lui.

2-

Placer Ia co pie du professeur dans un cartable (ou autre) devant vous de fa<;on a
ce que l'eleve ne voit pas ce que vous ecrivez.

3-

Dites ces directives a I'eleve:
Quand je dirai "vas-y"' tu pourras commencer a lire a voix haute du debut
du texte (montrez le debut avec votre doigt). Lis le texte du mieux que tu
peux. Je vais t'arreter apres 1 minute. Situ as de Ia difficulte avec un mot, je
t'aiderai. As-tudes questions?

4-

Dites le nom de I' eleve ou son numero de bibliotheque dans I' enregistreuse. Dites
ensuite: vas-y. Commencez le chronometre quand l'eleve dit le premier mot. Si
l'eleve hesite sur le premier mot du texte pendant 3 secondes, dites-lui le mot,
inscrivez-le comme une faute et commencez le chronometre quand il dira le
deuxieme mot.

5-

Suivez Ia lecture sur votre copie. Soulignez ou encerclez les mots que 1' eleve lit
incorrectement.

6-

Si l'eleve hesite ou ne sait pas un mot pendant 3 secondes, dites-lui le motet
inscrivez-le comme une faute.

7-

Ala fin de la minute, inscrivez une barre oblique(!) pour indiquer la fin de Ia
lecture et dites: arrete.

8-

Remerciez I'eleve et dites-lui un mot d'encouragement. Passez a l'eleve suivant.

* Occasionnellement, un eleve lira a toute vitesse, c'est-a-dire qu'illira rapidement et

sans expression. Dites-lui que ce n'est pas un test de vitesse et qu'il doit lire du mieux
qu'il peut. Recommencez la lecture.
·

51

Regles de correction pour CBM lecture
(Scoring Procedure)

1-

Mots Ius correctement

a)
b)
c)
d)
e)

Mots bien prononces
Mots corriges par le lecteur lui-meme
Mots repetes (comptes pour un mot)
Mots dits avec un accent different (dialecte)
Mots inseres sont ignores

2-

Mots consideres comme "incorrect"

a)
b)
c)
d)

Mots mal prononces ou changes pour un autre mot (meme si c' est une substitution
logique).
Mots omis
Hesitations
Mots dits dans le mauvais ordre (inverses)

3-

Regles speciales

a)
b)
c)
d)

Nombres numeraux et ordinaux
Mots avec apostrophes
Mots avec trait d'union
Abreviations

52

Appendix G
Letters of Permission
Information Letter to Parents

SCHOOL DISTRICT NO. 57 (PRINCE GEORGE)
1894 Ninth Avenue, Prince George, B.C. V2M 1L7

Phone: (250) 561-6800 Fax: (250) 561-6801
www .schdist57 .bc.ca

October 23, 2001
Sylvie St. Pierre
Teacher
College Heights Elementary School
Dear Sylvie:
This letter is to confirm our telephone conversation regarding your request to develop
reading and writing norms for the French Language Arts curriculum as a part of your
master's thesis.
You spoke with Dave Devore in September and received tentative permission to expand
the request for a project. Although the School District Curriculum Department is not
planning the development of French CBM norms in the next few years, we do give
permission in principle for this project to go forward. However, the permission of each of
the principals of the four schools mentioned will need to be obtained before you initiate
the project in their school.
Your letter of September 261h outlines the project very well. As we discussed during our
conversation, you have agreed to use the students' numbers rather than their names on
the data in order to protect confidentiality. Please also forward to me a copy of the
UNBC Ethics Committee approval.
The school district receives many requests for research projects. Although we do try to
support many of these projects, in principle, and encourage schools to accept research
students into their schools, we are unable to provide any release time for projects of this
nature.
If you have any questrons, please do not hesitate to call me. Good luck with your
project.
Sincerely,

~~

Bonnie Chappell
Director of Curriculum & Instruction
BC/hg

cc:

D. DeVore, Director
C. Anserello, School Services Administrator
French Immersion Principals

uflsc

UNIVERSITY OF NORTHERN
BRITISH COLUMBIA
Research Ethics Board

MEMORANDUM
To:

Sylvie St. Pierre
Peter MacMillan
Education Program

From :

Alex C. Michalos
Chair, Research Ethics Board

Date :

December 12, 2001

Re:

Ethics Proposal 2001.1207.112
Curriculum Based Measurement Norming Tables for Reading
Fluency and Written Expression for SO #57 French Immersion
Program

Thank you for submitting the above noted proposal to the UNBC Research Ethics Board
for review. Your proposal has been approved and you may begin your research .
If you have any questions regarding the above , please feel free to contact me.
Sincerely,
(

Dr. Alex C. Michalos, Chair
UNBC Research Ethics Board

L_ __ _

55
January 2002
Dear Parents,
During this school year, Learning Assistance Teachers (L.A.T.) in the French
Immersion schools are conducting a series of tests in reading and writing with French
Immersion students. These tests, Curriculum-Based Measurement (CBM), have been
used in our School District for many years for all the students at the elementary levels but
in English only. In a 1995-96 report, the School District has recommended the
development of probes and norming tables for French Immersion.
As a U.N.B.C. graduate student in Education, I have taken this endeavor on as the
final project for my Master of Education degree, and has been approved by the School
District Administration. Being a French Immersion Learning Assistance Teacher myself,
I am taking on a task that is part of the L.A.T.'s regular responsibility, thus developing a
project that is relevant for the students' ongoing evaluation. Although the results will be
part of the University project, both student names and school names will not be used in
the project nor in the report to the school board. Library numbers are used to keep track
of student data. However, each school L.A.T. will have access to student names and
results as assessing is part of their regular responsibility.
Testing is done in October, January and April for grade two to seven. Grade one
students are tested only in April. The series of three tests are to establish norming tables
so that educators can have a bench mark to compare students at different periods of the
school year. These tests are easy to.administer and measure the student progress in the
same curriculum used the French Immersion classroom. It is very difficult to find
standardized tests for French programs and the results of this project will help both
L.A.T.s and classroom teachers to develop programs relevant to the needs and skills of
the students.
If you have any concerns or questions about the tests or the UNBC project, you
can contact your school L.A.T. or myself, Sylvie St-Pierre, L.A.T. (UNBC student) at
College Heights Elementary School at 964-4408 or at home at 562-9268.
Merci, thank you!
I would like to reiterate and assure you that both school and student names are
kept confidential.
Sincerely,

Sylvie St-Pierre
French Immersion teacher grade 5/6
French Immersion Learning Assistance Teacher
Ecole College Heights Elementary School

56

Appendix H
Graph with Raw Scores and Graph with a Smoothed Curve
(Grade Six)

57

Words Read Correctly in Grade Six

130
120

I

110
100
90
80
0

a::
3:

•: .•.. ..

70
60

6

•

I el

.•

•

•• 6
•

•

6

•

....- ... • .-

6. ••••
6

•

u

-FaiiWRC

•

•

--.-Winter WRC
-springWRC

50
•
I 6 •
40 I • •
30
20
10
0
(;)

~~~~~~~~~~~~~~~~~~~~
Percentiles

Words Read Correctly in Grade Six
Smoothed

130
~
120 +---------------------------------------__.

... :.

110
100

~
~

~~ , ~

90
80
70 ~
60

~~

I •

• I
: •

50
~~~~~
40 I~

••

•

I

~~~ ~ ~

I •. •
~~~~
• ••
• .... ~ ~~
• ~~
-

~

~

~~~~~ ~~~~ ~ ~~ L ~L ~

• •

--.-Winter WRC

• •

~

30 -l--- - - - - - - ------------------------------------1
20
~
10 ~
0 ~

~
~,,

~

,

,

,, , , ~

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 10
Percentiles

-FaiiWRC
-springWRC

58

Appendix I
Descriptive Statistics of Written Expression (TWW) Results
for Three Norming Periods

59

Descriptive Statistics of Written Expression (TWW)

Grade 1
Grade 2
Grade 3
Grade 4
Grade 5
Grade 6
Grade 7

Fall
Winter
Spring
Fall
Winter
Spring
Fall
Winter
Spring
Fall
Winter
Spring
Fall
Winter
Spring
Fall
Winter
Spring
Fall
Winter
Spring

Mean

SD

15.48
11.82
17.41
19.12
20.23
25.20
24.86
28.43
35.26
36.53
35.12
40.97
44.77
42.62
.50.76
51.03
50.00
57.10
56.70

10.50
7.22
9.19
8.16
9.44
12.71
11.66
11.24
11.17
8.99
10.15
13.86
12.24
12.70
13.69
12.53
13.66
13.20
14.41

Min
1
1
4
5
5
8
8
2
6
21
8
10
24
19
27
28
30
35
30

Max
48
31
41
34
50
73
68
50
64
61
52
64
84
71
78
81
82
90
92

Skew
1.17
0.60
0.80
0.08
1.12 .
1.18
1.18
-0.15
0.02
0.51
-0.47
-0.43
0.83
0.38
0.31
0.50
0.25
0.65
0.22

Kurtosis
1.34
-0.39
0.03
-1.04
1.54
2.80
2.98
0.19
0.96
0.45
0.31
0.50
1.90
-0.25
-0.92
0.06
-0.75
0.01
-0.45

60

Appendix J
Norming Table for Total Words Written (TWW)
and Words Spelled Correctly (WSC) for Grade Six

61

Total Words Written and Words Spelled Correctly
Grade six
Fall

Winter

Spring

Percentile

TWW

wsc

TWW

wsc

TWW

wsc

99
95
90
85
80
75
70
65
60
55
50
45
40
35
30
25
20
15
10
5
1

70
66
63
56
51
50
49
45
43
42
42
40
39
38
37
35
32
29
28
25
20

64
62
53
49
47
45
44
43
40
38
38
37
35
32
29
28
27
26
23
22
16

77
71
66
66
63
59
55
52
51
50
49
47
44
43
40
39
37
36
34
31
28

71
68
62
58
54
53
50
48
47
46
45
42
40
38
36
34
33
31
29
24
19

80

76
70
62
59
55
54
51
49
48
47
46
44
42
40
38
36
34
32
31
28
24

72

67
66
63
59
56
53
52
51
50
49
48
46
45
43
41
39
38
32
29

Description

Well above average
Above Average

Average

Below Average

Well Below Average