STAR

Math

Rating Summary

Classification Accuracyfull bubble
GeneralizabilityBroad
Reliabilityfull bubble
Validityfull bubble
Disaggregated Reliability and Validity Datafull bubble
Efficiency
AdministrationIndividual Group
Administration & Scoring Time10 Minutes
Scoring KeyComputer Scored
Benchmarks / NormsYes
Cost Technology, Human Resources, and Accommodations for Special Needs Service and Support Purpose and Other Implementation Information Usage and Reporting

Cost per student for year 1:
$9.00 per student (based on start up, per student fees based on minimum of 203 students. $1,599.00 + $200.00 fee = $1799.00)
The one time fee for STAR Math is $1,599.00. It is web-based and includes the tool, software/technical manual, installation guide, testing instructions, remote installation and setup.

Cost for subsequent years:

$0.99 per student per year (203 student minimum).

Computer access is required.
Testers will require less than 1 hour of training.

Paraprofessionals can administer the test.

STAR Math is a computer-adaptive assessment, and the difficulty of items is adjusted automatically to reflect the skill level of a student, including students with special needs.

STAR Math offers several accommodations for students with disabilities through the accessibility options built into a computer’s operating system.

For students with limited vision, the introductory screens of STAR Math respond to the “high contrast” accessibility feature within Windows and the “switch to black and white” accessibility feature in MAC OS 10. The assessment screens within STAR Math already provide visual contrast through a yellow background and black writing. Furthermore, STAR Math is compatible with Mac OS 10’s “zoom in” accessibility feature, which allows users to magnify nearly all STAR Math screens.

Renaissance Learning, Inc.
PO Box 8036 Wisconsin Rapids, WI 54495
http://www.renaissance.com/
Phone: (800) 338-4204

Field tested training manuals are included and should provide all implementation information.

Ongoing technical support is available from the Renaissance Learning technical support staff.

STAR Math is a computer-adaptive assessment that allows teachers to assess students’ mathematics abilities accurately in 15 minutes or less.

STAR Math can be administered to individuals or groups of students in grades 1 – 12.

The tool provides information on students performance in mathematics. The difficulty of items is adjusted automatically to reflect the skill level of the student, including those students with special needs.

STAR Math helps educators accelerate learning and increase motivation by providing immediate individualized feedback on academic tasks and classroom achievement.

STAR Math takes approximately 15 minutes to administer the test. Scoring is automated by the software and does not require any additional time.

STAR Math uses a proprietary, Rasch-based 1-parameter logic response model to calculate scores.  The scaled score is a non-linear monotonic  transformation of the Rasch ability estimate resulting from the adaptive test.  STAR Math scaled scores range from 1-1,400.  From the scaled scores and the student’s current grade placement, are derived grade equivalent, percentile, and normal curve equivalent scores. Reports generated by STAR Math can include all of these scores.

The number of alternate forms is virtually unlimited because STAR Math is a computer adaptive assessment.

 

Classification Accuracy

Classification Accuracy in Predicting Proficiency on State Achievement Tests in 7 States (Arkansas, Delaware, Illinois, Kansas, Michigan, Mississippi, and North Carolina) and the Terra Nova in Oklahoma
n=29,594
False Positive Rate 0.26
False Negative Rate 0.25
Sensitivity 0.75
Specificity 0.74
Positive Predictive Power 0.47
Negative Predictive Power 0.91
Overall Classification Rate 0.75
AUC (ROC) Grade AUC
2          0.81
3          0.82
4          0.82
5          0.84
6          0.85
7          0.83
8          0.80
Base Rate 0.23
Cut Point: Grade Cut Score
2          421
3          523
4          607
5          658
6          708
7          744
8          759

 

Classification Accuracy in Predicting Proficiency on the South Dakota State Test of Educational Progress
n=27,000
False Positive Rate 0.21
False Negative Rate 0.26
Sensitivity 0.75
Specificity 0.79
Positive Predictive Power 0.59
Negative Predictive Power 0.88
Overall Classification Rate 0.77
AUC (ROC) Grade AUC
3          0.83
4          0.85
5          0.84
6          0.87
7          0.86
8          0.89
Base Rate 0.29
Cut Point: Grade Cut Score
3          499
4          579
5          642
6          705
7          741
8          786

 

Generalizability

Description of Study 1 Sample:

  • Number of States: 6 (Arkansas, Delaware, Illinois, Michigan, Mississippi, Kansas)
  • Size: 18,969
  • Region:
    • Central
    • Southwest
    • Northeast
    • Midwest
    • Southeast
  • Gender:
    • 30.9% Male
    • 31.4% Female
    • 37.7% Unknown
  • Race/Ethnicity:
    • 28.8% White, Non-Hispanic
    • 15.5% Black, Non-Hispanic
    • 2.4% Hispanic
    • 0.1% American Indian/Alaska Native
    • 0.4% Asian/Pacific Islander
    • 46.8% Unknown

Description of Study 2 Sample:

  • Number of States: 1 (South Dakota)
  • Size: 27,000
  • Gender:
    • 49.9% Male
    • 50.1% Female
  • Race/Ethnicity:
    • 63.3% White, Non-Hispanic
    • 0.7% Black, Non-Hispanic
    • 1.3% Hispanic
    • 33.3% American Indian/Alaska Native
    • 0.7% Asian/Pacific Islander

Cross Validation Study Description of study sample:

  • Number of States: 7 (Arkansas, Delaware, Illinois, Michigan, Mississippi, Kansas)
  • Size: 5,959
  • Region:
    • Central
    • Southwest
    • Northeast
    • Midwest
    • Southeast
  • Gender:
    • 33.9%  Male
    • 35.4%  Female
    • 30.7% Unknown
  • Race/Ethnicity:
    • 23.7% White, Non-Hispanic
    • 11.8% Black, Non-Hispanic
    • 2.4% Hispanic
    • 0.3% American Indian/Alaska Native
    • 0.3% Asian/Pacific Islander
    • 61.5% Unknown

Reliability

Type of Reliability Age or Grade n Coefficient SEM
Generic Grades 1 - 12 29,228 0.947 40
Split-half Grades 1 - 12 29,228 0.944  
Alternate forms/Test-retest Grades 1 - 12 7,389 0.908  

 

Validity

Type of Validity Age or Grade Test or Criterion n (range) Coefficient
Range Median
Concurrent Grades 3 – 8 Idaho Standards Achievement Test 2,458 (170 – 231) 0.68 – 0.85 0.80
Concurrent Grades 3 - 7 Michigan Educational Assessment Program 1,179 (53 – 162) 0.58 – 0.84 0.76
Concurrent Grades 2 – 6,  and 8 Delaware Student Testing Program 1,330 (44 – 2,960) 0.56 – 0.78 0.72
Concurrent Grades 3 and 5 Minnesota Comprehensive Assessment 340 (81 – 91) 0.71 – 0.76 0.74
Concurrent Grades 3 – 6 Mississippi Curriculum Test 442 (52 – 154) 0.43 – 0.78 0.74
Concurrent Grades 3 – 8 Mississippi Curriculum Test 2nd Edition 8,017 (572 – 1,909) 0.65 – 0.77 0.72
Concurrent Grades 2 – 5 TerraNova 1,314 (119 – 205) 0.45 – 0.78 0.71
Predictive Grades 3 – 7 Michigan Educational Assessment Program 840 (37 – 84) 0.63 – 0.87 0.78
Predictive Grades 3 – 5 Mississippi Curriculum Test 583 (33 – 164) 0.51 – 0.82 0.71
Predictive Grades 3 – 8 Texas Assessment of Knowledge and Skills 2,397 (135 – 646) 0.49 – 0.74 0.70
Predictive Grades 3 – 8 Mississippi Curriculum Test 2nd Edition 15,774 (2,148 – 2,977) 0.60 – 0.72 0.67

 

Disaggregated Reliability, Validity, and Classification Data for Diverse Populations

Classification Accuracy in Predicting Proficiency on State Achievement Tests in 7 States (Arkansas, Delaware, Illinois, Kansas, Michigan, Mississippi, and North Carolina) and the Terra Nova in Oklahoma: by Race/Ethnicity
  White, non-Hispanic
n = 5,665
Black, non-Hispanic
n = 2,914
Hispanic
n = 505
Asian/Pacific Islander
n = 68
American Indian/ Alaska Native
n = 99
False Positive Rate 0.2256 0.3677 0.2924 0.1455 0.1525
False Negative Rate 0.2600 0.1830 0.2147 0.1538 0.3500
Sensitivity 0.7400 0.8170 0.7853 0.8462 0.6500
Specificity 0.7744 0.6323 0.7076 0.8545 0.8475
Positive Predictive Power 0.4262 0.4811 0.5614 0.5789 0.7429
Negative Predictive Power 0.9293 0.8922 0.8736 0.9592 0.7813
Overall Classification Rate 0.7680 0.6867 0.7327 0.8529 0.7677
AUC (ROC) Grade  AUC
2          0.841
3          0.828
4          0.844
5          0.862
6          0.878
7          0.801
8          0.822
Grade  AUC
2          0.625
3          0.819
4          0.798
5          0.811
6          0.809
7          0.756
8          0.833
Grade  AUC
2          1.000
3          0.815
4          0.852
5          0.872
6          0.860
7          0.792
8          0.693
Grade  AUC
2          n/a
3          n/a
4          1.000
5          1.000
6          1.000
7          n/a
8          1.000
Grade  AUC
2          1.000
3          0.900
4          0.939
5          0.825
6          1.000
7          0.000
8          1.000
Base Rate 0.1846 0.2944 0.3228 0.1912 0.4040
Cut Score Grade Cut Score
2          421
3          523
4          607
5          658
6          708
7          744
8          759
Grade Cut Score
2          421
3          523
4          607
5          658
6          708
7          744
8          759
Grade Cut Score
2          421
3          523
4          607
5          658
6          708
7          744
8          759
Grade Cut Score
2          421
3          523
4          607
5          658
6          708
7          744
8          759
Grade Cut Score
2          421
3          523
4          607
5          658
6          708
7          744
8          759

Disaggregated Reliability

Type of Reliability Age or Grade n (range) Coefficient
Range Median
Generic (White) Grades 1 – 12 48,002 0.81 – 0.93 0.89
Generic (Black) Grades 1 – 12 21,806 0.93 – 0.93 0.90
Generic (Hispanic) Grades 1 – 12 28,231 0.81 – 0.93 0.90