Benchmark Statistics Data Import Format and Sample File

Benchmark Statistics Data Import Format

To create the CSV file, use the data format table and the sample csv file as a guide, located at the end of end of the article.

Benchmark Statistics Import Sample File 

 

Header

Required

Data Type

Data Requirements

ItemName

Yes

String

ItemName must match the Item Name for an existing item in the Blueprint.

SampleSize

No

Integer

The number of cases used in the analysis.

Validation:
  • > 0

DifficultyProportionCorrect

No

Float

Validation:
  • 0.0 to 1.0
  • Rounded to 9 decimals

CorrectedItemTotalCorrelation

No

Float

Validation:
  • -1.0 to 1.0
  • Rounded to 9 decimals

ItemTotalCorrelation

No

Float

Validation:
  • -1.0 to 1.0
  • Rounded to 9 decimals

StandardDeviation

No

Float

The standard deviation of the scored item responses.

Validation:
  • 0 to 0.5
  • Rounded to 9 decimals

CorrectedOptionTotalCorrelation_#

No

Float

The corrected option-total correlation for each option; the correlation between selecting the option and the adjusted total exam score.

"#" should be replaced with the Response Option ID, and there should be a column for each response option for the given item type.

Validation:
  • -1 to 0.5
  • Rounded to 9 decimals

CorrectedOptionTotalCorrelation_Missing

No

Float

The corrected option-total correlation for omitted/not reached missing responses; the correlation a missing response and the adjusted total exam score.

Validation:
  • -1 to 0.5
  • Rounded to 9 decimals

ProportionChoosing_#

No

Float

The proportion of the sample who selected the option.

"#" should be replaced with the Response Option ID, and there should be a column for each response option for the given item type.

Validation:
  • 0.0 to 1.0
  • Rounded to 9 decimals

ProportionChoosing_Missing

No

Float

The proportion of the sample who omitted or did not reach the item.

Validation:
  • 0.0 to 1.0
  • Rounded to 9 decimals

AngoffCoefficient

No

Float

The judged probability that a minimally-qualified candidate would pass this item.
     – Or --
EmpiricalAngoffCoefficient - EAC, the empirical probability that a minimally-qualified candidate would pass this item computed using borderline groups, or a similar methodology.

Validation:
  • 0.0 to 1.0
  • Rounded to 9 decimals

Comment

No

String

Comment about the Benchmark statistics, such as a description of the sample group.

Validation:
  • 500 characters or less

 

Contact Us

If you have any questions or need additional assistance, please contact us by either emailing support@certiverse.com or by submitting a ticket from this article.