edSpring / Edcheckup Standard Reading Passages

Oral Reading Fluency

Cost Technology, Human Resources, and Accommodations for Special Needs Service and Support Purpose and Other Implementation Information Usage and Reporting

edSpring: Contact for cost details.

Edcheckup: Each license (or “set”) allows you to access materials, enter data, and generate reports for up to 35 students.

The cost per set is as follows:

  • If purchasing 1 – 10 sets, $100 per set
  • If purchasing 11 – 20 sets, $95 per set
  • If purchasing 21 – 100 sets, $90 per set
  • If purchasing 101 – 200 sets, $85 per set
  • If purchasing 201+ sets, $8 per set
This includes all materials and full data/reporting functionality for all subjects.

Computer and Internet access are required for full use of product services.

Testers will require less than 1 hour of training.

Paraprofessionals can administer the test.
Where to obtain:
 
edSpring
TIES
1667 Snelling Ave. N
St. Paul, MN 55108
 
Edcheckup
Edcheckup, LLC
7701 York Ave S, Ste 250
Edina, MN 55435
Website:  www.edcheckup.com
 
edSpring / Edcheckup: Field-tested training manuals are included and should provide all implementation information.
Ongoing technical support is available.

Edcheckup Standard Reading Passages consist of 20 passages at the same difficulty level for use for repeated 1 minute samples of reading aloud from text.

Testers follow along on a duplicate passage while the student reads aloud, and mark incorrectly read works and coding errors if desired. Testers are encouraged to rate the degree of prosody evidence in the student’s oral reading of the passage. Raw scores are entered into the Edcheckup online system.

Edcheckup is individually administered and requires 2 minutes for administration and 1 additional minute for scoring.

Raw scores are available. 

20 alternate forms are available.

 

Reliability of the Performance Level Score

Grade123456
RatingFull bubbleFull bubbleFull bubbleFull bubbleFull bubbleFull bubble
Type of Reliability Age or Grade n (range) Coefficient SEM Information / Subjects
range median
Cronbach’s Alpha 1 89 0.9434-0.9495 0.9446 Median = 4.04 Alpha = 0.9554, Standardized Item Alpha = 0.9582
Cronbach’s Alpha 2 216 0.8990-0.9164 0.9016 Median = 2.56 Alpha = 0.9554 Standardized Item Alpha = 0.9582, 17.9% Students of Color, 7.6% Students of Poverty, 13.4% Students with Disabilities
Cronbach’s Alpha 3 227 0.9056-0.9152 0.9152 Median = 2.57 Alpha = 0.9554, Standardized Item Alpha = 0.9582, 19.7% Students of Color, 7.5% Students of Poverty, 15.1% Students with Disabilities
Cronbach’s Alpha 4 206 0.9048-0.9303 0.9204 Median = 2.53 Alpha = 0.9554, Standardized Item Alpha = 0.9582, 23.8% Students of Color, 6.0% Students of Poverty, 13.1% Students with Disabilities
Cronbach’s Alpha 5 255 0.8703-0.8965 0.8860 Median = 2.44 Alpha = 0.9554, Standardized Item Alpha = 0.9582, 21.8% Students of Color, 5.6% Students of Poverty, 11.6% Students with Disabilities
Cronbach’s Alpha 6 78 0.8930-0.9229 0.9046 Median = 4.61 Alpha = 0.9554, Standardized Item Alpha = 0.9582, 33.3% Students of Color, 16.7% Students of Poverty, 11.5% Students with Disabilities
Alternate Form 2 216 0.899-0.916 0.902   17.9% Students of Color, 7.6% Students of Poverty, 13.4% Students with Disabilities
Alternate Form 3 229 0.906-0.915 0.914   19.7% Students of Color, 7.5% Students of Poverty, 15.1% Students with Disabilities
Alternate Form 4 206 0.905-0.930 0.920   23.8% Students of Color, 6.0% Students of Poverty, 13.1% Students with Disabilities
Alternate Form 5 255 0.870-0.897 0.886   21.8% Students of Color, 5.6% Students of Poverty, 11.6% Students with Disabilities
Alternate Form 6 78 0.893-0.923 0.905   33.3% Students of Color, 16.7% Students of Poverty, 11.5% Students with Disabilities

Reliability of the Slope

Grade123456
RatingFull bubbleFull bubbleFull bubbleFull bubbleFull bubbleFull bubble

In our reliability of slope analysis we used Latent Growth Modeling (LGM) to estimate the reliability of longitudinal weekly growth data as demonstrated by Yeo, Kim, Branum-Martin, Wayman, and Espin (2011). These researchers point out, "The feature of LGM that treats time as an independent variable at each occasion makes it possible to estimate the reliability of longititudinal data such as CBM data" (p. 276). An important conclusion of their study was, “ a crucial advantage of LGM is that reliability estimated by LGM takes into account the developmental trajectories at the individual level of analysis (Tisak & Tisak, 2000)” (p. 287).

The reliabilities for 9 weekly assessments at each grade level, grades 1-6, are presented in the table below.  For each grade level (and the entire sample) we report sample sizes, means, model-implied observed score variance, measurement error variance, observed variance - error variance, SEM, reliability for each assessment, and the median reliability for each grade level.  The range of median reliability coefficients for weekly progress monitoring probes across the six grades levels was 0.73 to 0.89 with a median of 0.81.

Type of Reliability

Grade

Assessment

n

Mean

Model-implied observed score variance

Measure-ment error variance

Observed variance - error variance

SEM

Reliability

Latent Growth Modeling

all

Assess1

2029

74.44

1966.10

219.73

1746.37

14.82

0.89

Latent Growth Modeling

all

Assess2

2029

76.06

1907.83

195.29

1712.54

13.97

0.90

Latent Growth Modeling

all

Assess3

2029

78.09

1951.90

234.21

1717.69

15.30

0.88

Latent Growth Modeling

all

Assess4

2029

79.66

1927.61

213.60

1714.01

14.62

0.89

Latent Growth Modeling

all

Assess5

2029

81.43

1913.37

204.53

1708.84

14.30

0.89

Latent Growth Modeling

all

Assess6

2029

83.23

1924.56

206.26

1718.30

14.36

0.89

Latent Growth Modeling

all

Assess7

2029

85.21

1995.32

194.37

1800.95

13.94

0.90

Latent Growth Modeling

all

Assess8

2029

86.21

1926.07

206.16

1719.92

14.36

0.89

Latent Growth Modeling

all

Assess9

2029

87.58

1915.85

205.17

1710.67

14.32

0.89

Latent Growth Modeling

Median

 

 

 

 

 

 

 

0.89

 

 

 

 

 

 

 

 

 

 

Latent Growth Modeling

1

Assess1

267

23.98

302.22

72.28

229.94

8.50

0.76

Latent Growth Modeling

1

Assess2

267

25.19

325.45

69.83

255.62

8.36

0.79

Latent Growth Modeling

1

Assess3

267

27.26

356.97

72.74

284.23

8.53

0.80

Latent Growth Modeling

1

Assess4

267

29.66

416.99

72.34

344.65

8.51

0.83

Latent Growth Modeling

1

Assess5

267

31.50

442.08

87.55

354.53

9.36

0.80

Latent Growth Modeling

1

Assess6

267

33.23

459.59

71.49

388.10

8.46

0.84

Latent Growth Modeling

1

Assess7

267

35.24

497.72

81.67

416.05

9.04

0.84

Latent Growth Modeling

1

Assess8

267

36.63

525.61

99.11

426.50

9.96

0.81

Latent Growth Modeling

1

Assess9

267

39.96

674.19

113.48

560.71

10.65

0.83

Latent Growth Modeling

Median

 

 

 

 

 

 

 

0.81

 

 

 

 

 

 

 

 

 

 

Latent Growth Modeling

2

Assess1

412

41.77

818.61

215.78

602.83

14.69

0.74

Latent Growth Modeling

2

Assess2

412

43.77

809.76

126.18

683.58

11.23

0.84

Latent Growth Modeling

2

Assess3

412

46.20

836.64

130.46

706.17

11.42

0.84

Latent Growth Modeling

2

Assess4

412

47.94

752.42

127.84

624.58

11.31

0.83

Latent Growth Modeling

2

Assess5

412

49.13

846.01

129.77

716.24

11.39

0.85

Latent Growth Modeling

2

Assess6

412

51.47

813.24

123.38

689.87

11.11

0.85

Latent Growth Modeling

2

Assess7

412

53.57

906.92

101.97

804.95

10.10

0.89

Latent Growth Modeling

2

Assess8

412

55.70

1004.91

202.09

802.83

14.22

0.80

Latent Growth Modeling

2

Assess9

412

56.52

896.76

144.52

752.25

12.02

0.84

Latent Growth Modeling

Median

 

 

 

 

 

 

 

0.84

 

 

 

 

 

 

 

 

 

 

Latent Growth Modeling

3

Assess1

491

77.50

1202.54

138.87

1063.67

11.78

0.88

Latent Growth Modeling

3

Assess2

491

80.83

1188.67

133.77

1054.90

11.57

0.89

Latent Growth Modeling

3

Assess3

491

83.36

1324.91

224.67

1100.24

14.99

0.83

Latent Growth Modeling

3

Assess4

491

86.53

1383.40

311.23

1072.18

17.64

0.78

Latent Growth Modeling

3

Assess5

491

87.33

1204.38

141.96

1062.43

11.91

0.88

Latent Growth Modeling

3

Assess6

491

88.90

1270.41

133.28

1137.13

11.54

0.90

Latent Growth Modeling

3

Assess7

491

91.46

1459.82

224.46

1235.35

14.98

0.85

Latent Growth Modeling

3

Assess8

491

92.66

1273.58

137.24

1136.34

11.72

0.89

Latent Growth Modeling

Median

Assess9

491

93.27

1261.91

162.22

1099.70

12.74

0.87

 

 

 

 

 

 

 

 

 

0.88

 

 

 

 

 

 

 

 

 

 

Latent Growth Modeling

4

Assess1

352

98.46

1365.94

481.32

884.63

21.94

0.65

Latent Growth Modeling

4

Assess2

352

98.76

1087.14

273.45

813.69

16.54

0.75

Latent Growth Modeling

4

Assess3

352

100.56

1056.35

257.96

798.39

16.06

0.76

Latent Growth Modeling

4

Assess4

352

100.13

1087.95

198.58

889.37

14.09

0.82

Latent Growth Modeling

4

Assess5

352

103.21

1022.30

207.64

814.66

14.41

0.80

Latent Growth Modeling

4

Assess6

352

105.81

1205.94

276.01

929.93

16.61

0.77

Latent Growth Modeling

4

Assess7

352

106.20

1161.25

165.32

995.93

12.86

0.86

Latent Growth Modeling

4

Assess8

352

107.14

1091.65

232.15

859.50

15.24

0.79

Latent Growth Modeling

4

Assess9

352

107.82

1207.99

207.29

1000.70

14.40

0.83

Latent Growth Modeling

Median

 

 

 

 

 

 

 

0.79

 

 

 

 

 

 

 

 

 

 

Latent Growth Modeling

5

Assess1

329

103.12

1227.87

301.66

926.21

17.37

0.75

Latent Growth Modeling

5

Assess2

329

104.56

1165.03

305.33

859.71

17.47

0.74

Latent Growth Modeling

5

Assess3

329

107.04

1301.86

451.24

850.62

21.24

0.65

Latent Growth Modeling

5

Assess4

329

106.05

1279.40

300.96

978.44

17.35

0.76

Latent Growth Modeling

5

Assess5

329

106.95

1153.05

285.32

867.73

16.89

0.75

Latent Growth Modeling

5

Assess6

329

110.37

1179.15

319.27

859.88

17.87

0.73

Latent Growth Modeling

5

Assess7

329

113.45

1177.76

318.17

859.59

17.84

0.73

Latent Growth Modeling

5

Assess8

329

111.49

1227.17

350.06

877.11

18.71

0.71

Latent Growth Modeling

5

Assess9

329

113.66

1177.98

326.42

851.56

18.07

0.72

Latent Growth Modeling

Median

 

 

 

 

 

 

 

0.73

 

 

 

 

 

 

 

 

 

 

Latent Growth Modeling

6

Assess1

178

116.90

1507.86

344.73

1163.12

18.57

0.77

Latent Growth Modeling

6

Assess2

178

116.55

1292.55

338.50

954.06

18.40

0.74

Latent Growth Modeling

6

Assess3

178

115.99

1342.37

280.82

1061.55

16.76

0.79

Latent Growth Modeling

6

Assess4

178

120.14

1317.25

242.36

1074.89

15.57

0.82

Latent Growth Modeling

6

Assess5

178

123.16

1317.93

279.80

1038.14

16.73

0.79

Latent Growth Modeling

6

Assess6

178

121.14

1327.09

454.85

872.24

21.33

0.66

Latent Growth Modeling

6

Assess7

178

124.65

1318.15

314.48

1003.67

17.73

0.76

Latent Growth Modeling

6

Assess8

178

125.46

1376.98

251.33

1125.65

15.85

0.82

Latent Growth Modeling

6

Assess9

178

127.28

1327.66

355.22

972.44

18.85

0.73

Latent Growth Modeling

Median

 

 

 

 

 

 

 

0.77

Information (including normative data) / Subjects:

Total students in subject sample was 2,029. The percentage of students that were American Indian was 5.3%, for African American was 29.7%, for Asian American was 9.9%, for Hispanic American was 14.4%, and for White American was 40.1%. 47% of the sample was female. 18.8% of the sample were students with disabilities. 

Validity of the Performance Level Score

Grade123456
RatingFull bubbleFull bubbleFull bubbleFull bubbleFull bubbleFull bubble
Type of Validity Age or Grade Test or Criterion n (range) Coefficient Information / Subjects
range median
Construct 2 Measure of Academic Progress 215   0.756 16.6% Students of Color, 7.1%, Students of Poverty, 13.3% Students with Disabilities
Construct 3 Measure of Academic Progress 229   0.626 19.5% Students of Color, 7.1%, Students of Poverty, 14.2% Students with Disabilities
Construct 4 Measure of Academic Progress 206   0.694 22.3% Students of Color, 5.9%, Students of Poverty, 11.9% Students with Disabilities
Construct 5 Measure of Academic Progress 256   0.670 20.4% Students of Color, 5.4%, Students of Poverty, 11.2% Students with Disabilities
Construct 6 Measure of Academic Progress 78   0.760 34.7% Students of Color, 17.3%, Students of Poverty, 10.7% Students with Disabilities
Predictive 2 Measure of Academic Progress 204   0.764 16.4% Students of Color, 5.8%, Students of Poverty, 12.6% Students with Disabilities
Predictive 3 Measure of Academic Progress 219   0.669 18.7% Students of Color, 5.9%, Students of Poverty, 14.2% Students with Disabilities
Predictive 4 Measure of Academic Progress 195   0.661 22.1% Students of Color, 5.1%, Students of Poverty, 11.3% Students with Disabilities
Predictive 5 Measure of Academic Progress 157   0.583 17.2% Students of Color, 3.2%, Students of Poverty, 8.3% Students with Disabilities

Predictive

1 Maze Comprehension 245   0.68  

Predictive

6 Maze Comprehension 158   0.61  

 

Predictive Validity of the Slope of Improvement

Grade123456
RatingFull bubbleFull bubbleEmpty bubbleFull bubbledashdash
Type of Validity Age or Grade Test or Criterion n (range) Coefficient Information (including normative data)/Subjects
range median
Predictive K, 1, 3 NAEP-MAP 37   0.57 Heirarchical multiple regression analysis using Oral Reading growth from Spring of Kindergarten to Fall of 1st grade to predict end of 3rd grade performance on NAEP-MAP where (AF (1,33) = 8.48, p < 0.01) and accounted for 11% more of the variance.
Predictive K, 1, 3 NAEP-MAP 37   0.68 Heirarchical multiple regression analysis using Oral Reading growth from Spring of Kindergarten to Spring of 1st grade to predict end of 3rd grade performance on NAEP-MAP where (AF (1,33) = 22.51, p < 0.001) and accounted for 22% more of the variance.
Predictive 1, 3 NAEP-MAP 37   0.69 Heirarchical multiple regression analysis using Oral Reading growth from Fall of 1st grade to Spring of 1st grade to predict end of 3rd grade performance on NAEP-MAP where (AF (1,33) = 13.15, p < 0.01) and accounted for 13% more of the variance.

Two studies are described below that provide evidence bearing on this GOM.

Study 1
A special type of the structural equation modeling approach, latent growth modeling (LGM) was used to the proposed growth model. A conditional LGM with MAP scores was employed to investigate whether CBM reading aloud slope predicts performance on MAP scores, after controlling for initial status on CBM reading aloud (See figure 1).
                      Figure 1.  Latent growth modeling with three time points
Figure 1.  Latent growth modeling with three time points

In evaluating the overall goodness of fit, the comparative fit index (CFI), normed fit index (NFI), and incremental fit index (IFI) were used in the study. Models leading to a CFI, NFI, and IFI higher 0.90 were considered a reasonable good fit. In the table below we present sample means and standard deviations for the study variables.

  Kindergarten Fall Spring MAP
Mean 26.7 41.2 90.2 210.0
S.D 29.7 35.0 44.1 10.8

Pearson correlations were calculated to provide information on the rank-ordered reliability and validity of CBM data and shown in the table below.

  Kindergarten Fall Spring
Kindergarten 1    
Fall 0.94 1  
Spring 0.77 0.83 1
MAP 0.68 0.75 0.82

To examine the value of the CBM growth rate as a significant predictor of improved MAP reading scores, the conditional LGM with MAP was used in this study. All fit measures, including IFI, CFI, and NFI indicated a fit within the acceptable to good range. The parameter estimates (slope and intercept) are presented in the table below.

  Estimate SE
Slope 63.41*** 4.80
Intercept 26.83*** 5.24

*** p<0.001

The estimated slope was 63.41 WRCM/month and the estimated initial status was 26.84 WRCM. The table below provides the direct effect coefficient of intercept and slope on MAP.

  Intercept-> MAP Slope-> MAP
Estimate 0.22***    0.19***
SE 0.04 0.05

*** p<0.001

According to the results the direct effect of the CBM slope was significant, meaning that increasing growth rate predicted performance on MAP over time, after controlling for initial status on CBM reading aloud (Kindergarten). In addition, the intercept (initial status) of CBM reading aloud contributed statistically to the prediction of MAP performance, after controlling for the slope of CBM reading aloud. In conclusion, both the intercept and slope of CBM reading aloud were significant predictors of performance on MAP scores.

Study 2 (HLM)
For these analyses the OR dataset included 528 students with 10 or more measures for OR where at least 14 days had passed between the fourth to last and the third to last measures.

With this sample, HLM was used to estimate individual student slopes across all but the last three measures. A separate HLM was run for each grade. These slopes were then used in a regression model to predict the average of the last three time points, called AVG, again by grade. Since fewer than 30 students met the criteria in Grades 5 & 6, the regression model was not used at those grade levels. Results from the regression analyses are included in the table below.

OR Regression Results for Slopes predicting AVG by Grade

Grade N R F p-value
1 221 0.81 431.50 0.000
2 137 0.62 85.72 0.000
3 92 0.21 4.24 0.042
4 57 0.91 267.50 0.000
5 16 - - -
6 5 - - -

 

Bias Analysis Conducted

Grade123456
RatingYesYesYesYesYesYes

Have you conducted additional analyses related to the extent to which your tool is or is not biased against subgroups (e.g., race/ethnicity, gender, socioeconomic status, students with disabilities, English language learners)?

Bias Analysis Method: Calculated Comparative Fit Indices, Tucker-Lewis-Indices, Root Mean Square Error of Approximation

Subgroups: White, African American, Asian, Hispanic, American Indian

Bias Analysis Results: See below

Test bias (measurement invariance) analysis results using invariance testing in multiple-group confirmatory factor models.

 

1. Model fit indices for each ethnic group

Ethnic group (n)

Chi-square* (df)

CFI

TLI

RMSEA (90% CI)

White (648)

96.541 (27)

0.993

0.991

0.063 (0.050-0.077)

African American (646)

99.024 (27)

0.993

0.990

0.064 (0.051-0.078)

Asian (192)

39.131 (27)

0.996

0.995

0.048 (0.000-0.080)

Hispanic (50)

38.578 (27)

0.986

0.981

0.093 (0.000-0.154)

American Indian (53)

42.681 (27)

0.981

0.975

0.105 (0.035-0.162)

 

* Chi-square values are influenced by sample size.

Overall, every ethnic group’s model-fit well. All relative fit indices (CFI, TLI) were above 0.98, which indicates very good model fit, given that a value over 0.90 is suggested to be a reasonably good fit for CFI and TLI (Hu & Bentler, 1998). In addition, the 90 percent confidence interval for RMSEA indicates very good model fit, given that a value below or at 0.05 is considered a very close approximate fit (Browne & Cudeck, 1993). These results show that all CBM-R assessments appeared to function equally well as one factor for all ethnic groups. 

Below are results for the measurement invariance (test bias) among different ethnic groups.

 

2. Model fit statistics for tests of structural/measurement invariance between White and other groups

(a) White versus African American

Model

Chi-square

df

Delta Chi-square

Delta df

Critical value (α= .05)

CFI

RMSEA

(90% CI)

Configural model

(equal structure; baseline)

195.565

54

-

-

-

0.993

0.064

(0.054-0.073)

Full metric model

(equal loadings)

212.670

62

17.105

8

15.507

0.992

0.061

(0.052-0.070)

Partial metric model* (equal loadings but one loading)

204.222

61

8.657

7

14.067

0.993

0.060

(0.051-0.069)

Full scalar model

(equal intercepts)

222.125

70

17.903

9

16.919

0.992

0.058

(0.049-0.067)

Partial scalar model

(equal intercepts but one intercept)

212.284

68

8.062

7

14.067

0.993

0.057

(0.049-0.066)

* Partial metric/scalar models were applied based on the model modification indices.

 

(b) White versus Asian

Model

Chi-square

df

Delta Chi-square

Delta df

Critical value

CFI

RMSEA

(90% CI)

Configural model

(equal structure; baseline)

135.672

54

-

-

-

0.994

0.060

(0.047-0.073)

Full metric model

(equal loadings)

151.275

62

15.603

8

15.507

0.993

0.059

(0.047-0.070)

Partial metric model* (equal loadings but one loading)

144.606

61

8.934

7

14.067

0.994

0.057

(0.045-0.069)

Full scalar model

(equal intercepts)

165.097

70

20.491

9

16.919

0.993

0.057

(0.046-0.068)

Partial scalar model

(equal intercepts but one intercept)

155.250

68

10.644

7

14.067

0.993

0.055

(0.044-0.067)

 

(c) White versus Hispanic

Model

Chi-square

df

Delta Chi-square

Delta df

Critical value

CFI

RMSEA

(90% CI)

Configural model

(equal structure; baseline)

135.119

54

-

-

-

0.992

0.066

(0.052-0.080)

Full metric model

(equal loadings)

154.734

62

19.615

8

15.507

0.991

0.065

(0.053-0.078)

Partial metric model* (equal loadings but one loading)

137.729

60

2.61

6

12.592

0.993

0.061

(0.048-0.074)

Full scalar model

(equal intercepts)

165.019

70

27.29

10

18.307

0.991

0.062

(0.050-0.075)

Partial scalar model

(equal intercepts but one intercept)

145.049

66

7.32

6

12.592

0.993

0.059

(0.046-0.072)

 

(d) White versus American Indian

Model

Chi-square

df

Delta Chi-square

Delta df

Critical value

CFI

RMSEA

(90% CI)

Configural model

(equal structure; baseline)

139.222

54

-

-

-

0.992

0.067

(0.053-0.081)

Full metric model

(equal loadings)

146.153

62

6.931

8

15.507

0.992

0.062

(0.049-0.075)

Partial metric model* (equal loadings but one loading)

N/A

N/A

N/A

N/A

N/A

N/A

N/A

Full scalar model

(equal intercepts)

155.698

70

9.545

8

15.507

0.992

0.059

(0.047-0.072)

Partial scalar model

(equal intercepts but one intercept)

N/A

N/A

N/A

N/A

N/A

N/A

N/A

 

For the first three comparisons (White vs. African American/Asian/Hispanic), the configural (equal structure) model had a good fit, and thus a series of model constraints were then applied in successive models to examine potential decrease in model fit. Two nested models were compared in model fit based on chi-square and degree of freedom differences at α= 0.05. The full metric invariance model fit well; but resulted in a marginally significant decrease in fit (chi-square difference) relative to the configural model. The partial metric invariance model (freeing just one loading), however, did not result in a significant decrease in fit, which indicates that CBM-R measures seem to be related to the factor equivalently between White and other groups. The full scalar model fit well; but also resulted in a marginally significant decrease in fit relative to the partial metric model. However, the partial scalar model (freeing just one intercept) did not result in a significant decrease in fit, which means that both ethnic groups (with the same level of reading) are expected to perform on CBM-R assessments in a very similar way. In case of the comparison between White and American Indian, both the full metric invariance model and the full scalar model did not result in a significant decrease in fit, which indicates that the same latent factor was being measured for the two groups and they have exactly the same expected performance on CBM-R if their levels are the same.

* Comparison between White and Hawaiian was not analyzed due to too small sample size of Hawaiian (n=4).

 

3. Conclusion

These results suggest that the same structure (configural invariance) and the same latent factor (full/partial metric invariance) were being measured in each ethnic group. In addition, the fact that the partial or full scalar invariance held indicates that White and other ethnic groups (with the same level of reading) have almost the same expected performance on CBM-R. In conclusion, there was no evidence on structural/measurement invariance (i.e., test bias) between White and other ethnic groups, meaning that CBM-R appears to be an unbiased measure among different ethnic groups.

Disaggregated Reliability and Validity Data

Grade123456
RatingNoYesYesYesYesYes

Disaggregated Reliability of the Performance Level Score

Type of Reliability Age or Grade n (range) Coefficient SEM Information / Subjects
range median
Cronbach’s Alpha 2 19 0.9554-0.9658 0.9633   African American
Alpha = 0.9844
Standardized item Alpha = 0.9868
Cronbach’s Alpha 3 20 0.8162-0.8811 0.8173   African American
Alpha = 0.9382
Standardized item Alpha = 0.9396
Cronbach’s Alpha 4 25 0.8720-0.9018 0.8989   African American
Alpha = 0.9543
Standardized item Alpha = 0.9608
Cronbach’s Alpha 5 25 0.9011-0.9289 0.9241   African American
Alpha = 0.9671
Standardized item Alpha = 0.9711
Cronbach’s Alpha 6 9 0.9749-0.9899 0.9825   African American
Alpha = 0.9929
Standardized item Alpha = 0.9941

Alternate Form

2 19 0.955-0.966 0.963   African American students in grade 2

Alternate Form

3 20 0.816-0.881 0.817   African American students in grade 3

Alternate Form

4 25 0.872-0.902 0.899   African American students in grade 4

Alternate Form

5 25 0.901-0.929 0.924   African American students in grade 5

Alternate Form

6 9 0.975-0.990 0.983   African American students in grade 6

Disaggregated Validity of the Performance Level Score

Type of Validity Age or Grade Test or Criterion n (range) Coefficient Information / Subjects
range median
Construct 2 Measure of Academic Progress 18   0.854 African American students
Construct 3 Measure of Academic Progress 20   0.774 African American students
Construct 4 Measure of Academic Progress 25   0.688 African American students
Construct 5 Measure of Academic Progress 25   0.580 African American students
Predictive 2 Measure of Academic Progress 16   0.831 African American students
Predictive 3 Measure of Academic Progress 20   0.665 African American students
Predictive 4 Measure of Academic Progress 24   0.689 African American students
Predictive 5 Measure of Academic Progress 15   0.775 African American students

 

Alternate Forms

Grade123456
RatingFull bubbleFull bubbleFull bubbleFull bubbleFull bubbleFull bubble

1. Evidence that alternate forms are of equal and controlled difficulty or, if IRT based, evidence of item or ability invariance:

Evidence of controlled difficulty for the passages used for progress monitoring at each grade was determined through use of the Flesch-Kinkaid Grade Level readability formula and the Lexile Framework for Reading (Metametrics, 2002). Both analyses indicate the passages were placed at appropriate grade levels. At grade 1 the passages had an average Flesch readability of 1.6 and Lexile of 272. At grade 2 the average Flesch readability was 2.4 and Lexile of 397. At grade 3 the average Flesch score was 3.3 and Flesch was 511. At grade 4 the average readability was 4.2 and Lexile was 656. The grade 5 Flesch readability was 5.8 and Lexile score was 759. Grade 6 Flesch readability was 6.9 and Lexile was 850.

Readability estimates for each probe (20) at each grade level (6). For each passage we list below the Lexile rating, as scored by Metametrics, and the Flesch readability score.

Grade Passage Lexile Flesch   Grade Passage Lexile Flesch
1 1 140 1.3   2 1 360 2.9
1 2 180 2.5   2 2 250 2.7
1 3 190 0.4   2 3 220 3.2
1 4 200 2.4   2 4 260 3.0
1 5 210 1.1   2 5 280 2.9
1 6 220 2.2   2 6 290 2.9
1 7 250 2.4   2 7 310 2.9
1 8 260 2.5   2 8 310 2.6
1 9 270 2.1   2 9 320 2.9
1 10 300 2.1   2 10 330 2.9
1 11 320 2.5   2 11 340 3.2
1 12 330 2.4   2 12 350 2.9
1 13 330 2.4   2 13 380 2.9
1 14 340 2.2   2 14 380 2.9
1 15 350 2.0   2 15 400 3.0
1 16 360 2.1   2 16 410 3.1
1 17 380 2.3   2 17 340 3.0
1 18 390 1.9   2 18 470 2.9
1 19 430 1.0   2 19 400 2.6
1 20 280 0.8   2 20 370 2.9

Grade Passage Lexile Flesch   Grade Passage Lexile Flesch
3 1 400 3.3   4 1 410 4.4
3 2 420 3.3   4 2 580 3.9
2 3 320 3.3   4 3 760 6.1
3 4 340 3.9   4 4 530 4.3
3 5 420 3.9   4 5 510 4.2
3 6 460 3.5   4 6 530 4.3
3 7 480 3.8   4 7 600 4.7
3 8 490 3.7   4 8 640 4.3
3 9 520 3.3   4 9 640 4.7
3 10 520 3.8   4 10 650 4.7
3 11 530 3.5   4 11 680 4.2
3 12 540 4.0   4 12 680 4.2
3 13 550 4.0   4 13 710 5.4
3 14 560 3.7   4 14 740 4.7
3 15 580 3.8   4 15 760 4.6
3 16 600 4.0   4 16 780 4.6
3 17 610 3.5   4 17 390 4.8
3 18 630 4.0   4 18 580 4.8
3 19 720 3.7   4 19 360 4.8
3 20 760 3.1   4 20 470 4.8

Grade Passage Lexile Flesch   Grade Passage Lexile Flesch
5 1 570 5.3   6 1 810 7.7
5 2 620 5.5   6 2 790 7.5
5 3 630 5.3   6 3 800 7.5
5 4 670 5.2   6 4 850 7.0
5 5 670 5.7   6 5 880 6.9
5 6 470 5.1   6 6 910 7.0
5 7 690 6.3   6 7 750 6.4
5 8 700 5.1   6 8 920 7.5
5 9 750 5.6   6 9 920 7.4
5 10 790 6.0   6 10 950 6.8
5 11 790 5.9   6 11 950 6.4
5 12 850 6.3   6 12 990 6.5
5 13 860 5.9   6 13 1060 7.5
5 14 670 5.2   6 14 770 6.1
5 15 970 5.7   6 15 470 6.5
5 16 1020 6.5   6 16 550 6.9
5 17 810 5.2   6 17 610 6.1
5 18 520 5.2   6 18 600 6.2
5 19 630 5.3   6 19 640 6.0
5 20 580 5.1   6 20 950 7.5

2. Number of alternate forms of equal and controlled difficulty:

20

Rates of Improvement Specified

Grade123456
RatingFull bubbleFull bubbleFull bubbleFull bubbleFull bubbleFull bubble

Is minimum acceptable growth (slope of improvement or average weekly increase in score by grade level) specified in your manual or published materials?

Yes

Specify the growth standards: 

Our growth standards are derived from a sample of 2007 students enrolled in grades 1 to 6 who were progress monitored on a weekly basis during the school year. The sample included 228 students in grade 1; 418 students in grade 2; 493 students in grades 3; 353 students in grade 4; 332 students in grade 5; and 183 students in grade 6. The average number of progress monitoring passages administered to these students was 16.5 (SD=6.3).

To specify "rates of improvement" for Words Read Correctly we calculated the slope for each student. We then developed percentiles for Words Read Correctly for each grade level. In the table below we provide the 25th, 50, and 75th percentiles at each grade, 1 through 6 (see table below).   Slopes associated with the 75th percentile are 1.6 at grade 1; 1.6 at grade 2; 1.4 at grade 3; 1.2 at grade 4; 1.4 at grade 5; and 1.3 at grade 6.

Grade

Percentile

Slope

Grade 1

N=228

25th

0.109

 

50th

0.647

 

75th

1.607

Grade 2

N=415

25th

0.616

 

50th

1.025

 

75th

1.570

Grade 3

N=493

25th

0.565

 

50th

0.948

 

75th

1.445

Grade 4

N=353

25th

0.336

 

50th

0.800

 

75th

1.194

Grade 5

N=332

25th

0.260

 

50th

0.833

 

75th

1.437

Grade 6

N=183

25th

0.218

 

50th

0.688

 

75th

1.310

 

What is the basis for specifying minimum acceptable growth?

Norm-Referenced

Basis for specifying minimum acceptable growth:

The Progress Monitoring system is structured so that teachers can choose from among 3 different growth rates at each grade level to set appropriate goals for their students. Those growth rates (termed “Modest,” “Reasonable”, and “Ambitious”) allow the teachers to choose growth rates for individual students that are consistent with their screening levels of performance and with the teacher’s knowledge of the student. The choice of growth rate then becomes a multiplier for individual goal setting and for setting up a progress graph that becomes a realistic basis for subsequent instructional decision making and program modifications. Once the student’s progress graph has been created, feedback on the degree of acceptability of student progress is provided by displaying the student’s scores in 3 color-coded ranges of risk exceeding desired growth, at desired growth, below desired growth. Developers used growth rates derived from a range of empirical research articles as a basis for establishing the desired growth rates for each grade level including the following:

Deno, S. L. & Marston, D. E. (2006) Curriculum-based Measurement of oral reading growth: An approach to measuring “fluency?” In Jay Samuels & Alan Dershwitz (Eds) Fluency International Reading Association: Newark, DE. 

Deno, S. L., Fuchs, L. S., Marston, D.B & Shin, J., (2001). Using curriculum-based measurement to establish growth standards for students with learning disabilities. School Psycholgy Review,30(4), 507-524.

Fuchs, L. S., Fuchs, D., Hamlett, C. L., Walz, L., & Germann, G. (1993). Formative evaluation of academic progress: How much growth can we expect? School Psychology Review, 22(1), 27-48.

Marston, D. & Magnusson, D. (1988) Curriculum-based measurement: District level implementation. In J. Graden, J. Zins, & M. Curtis (Eds.) Alternative educational delivery systems: Enhancing instructional options for all students. (pp 137-172)Washington, D.C.: National Association of School Psychologists.

Representation: Local    

Date: 2014-15 school year

Number of States: One state: Minnesota

Size: 2007 students

Gender: 53.1% Male; 46.9% Female

SES Indicators: 72.2 % of sample receives free or reduced lunch (FRL)

Race/Ethnicity: 40.1% White, Non-Hispanic; 29.7% Black, Non-Hispanic; 5.3% American Indian/Alaska Native; 9.9% Asian/Pacific Islander; 14.4% Hispanic

Disability classification: 18.8% students with disabilities

End-of-Year Benchmarks

Grade123456
RatingFull bubbleFull bubbleFull bubbleFull bubbleFull bubbleFull bubble

1. Are benchmarks for minimum acceptable end-of-year performance specified in your manual or published materials?

Yes.

a. Specify the end-of-year performance standards:

b. Basis for specifying minimum acceptable end-of-year performance:

Norm-referenced/Criterion-referenced

c. Specify the benchmarks:

The following year end Benchmarks are recommended in the Edcheckup manual.

Grade 1: 60 Words Read Correctly
Grade 2: 90 Words Read Correctly
Grade 3: 120 Words Read Correctly
Grade 4: 130 Words Read Correctly
Grade 5: 140 Words Read Correctly
Grade 6: 150 Words Read Correctly

d. Basis for specifying these benchmarks?

We think these are supportable benchmarks based on two separate studies with our passages. Our first study consists of data already analyzed for GOM 2. In that study the 50th percentile at each grade level was consistent with the benchmarks reported above. Grade 1: Median = 58.6, SD = 35.8, N = 746; Grade 2: Median = 100.0, SD = 37.5, N = 525; Grade 3: Median = 116.0, SD = 37.8, N = 637; Grade 4: Median = 119.0, SD = 37.2, N = 478; Grade 5: Median = 134.8, SD = 42.8, N = 382; Grade 6: Median = 149.0, SD = 50.0, N = 164.

In addition to the data in this analysis we provided, the following means were obtained in of our Edcheckup data when submitting the Standard Protocol for screening purposes. Grade 1: Mean = 73.9, SD = 37.2, N =90; Grade 2: Mean = 121.5, SD = 36.3, N = 216; Grade 3: Mean = 134.0, SD = 38.1, N = 229; Grade 4: Mean = 132.5, SD = 35.2, N = 207; Grade 5: Mean = 154.4, SD = 38.2, N = 256; Grade 6: Mean = 146.8, SD = 40.7, N = 78.

In selecting these benchmarks we not only used results from these two studies but also the large literature on oral reading scores that is available.

Sensitive to Student Improvement

Grade123456
RatingFull bubbleFull bubbleFull bubbleFull bubbleFull bubbleFull bubble

1. Describe evidence that the monitoring system produces data that are sensitive to student improvement (i.e., when student learning actually occurs, student performance on the monitoring tool increases on average).

Evidence of the sensitivity to student improvement is proved by the weekly growth rates for Edcheckup oral reading passages for grades 1 to 6 is below. These growth rates are comparable to those currently available in the literature on this topic.

Grade Sample Size Weekly Grown Rate
1 N = 421 1.52
2  N = 482 1.47
3 N = 573 1.36
4 N = 421 1.02
5 N = 349 1.02
6 N = 126 0.76

 

Decision Rules for Changing Instruction

Grade123456
Ratingdashdashdashdashdashdash

Decision Rules for Increasing Goals

Grade123456
Ratingdashdashdashdashdashdash

Improved Student Achievement

Grade123456
Ratingdashdashdashdashdashdash

Improved Teacher Planning

Grade123456
Ratingdashdashdashdashdashdash