testable_workshop_best_evaluation Class Reference

Inheritance diagram for testable_workshop_best_evaluation:

Collaboration diagram for testable_workshop_best_evaluation:

Public Member Functions
	normalize_grades (array $assessments, array $diminfo)
	average_assessment (array $assessments)
	weighted_variance (array $assessments)
	assessments_distance (stdclass $assessment, stdclass $referential, array $diminfo, stdclass $settings)

---

Detailed Description

Test subclass that makes all the protected methods we want to test public.

Definition at line 40 of file testlib.php.

---

Member Function Documentation

assessments_distance	(	stdclass $	assessment,
		stdclass $	referential,
		array $	diminfo,
		stdclass $	settings
	)

Measures the distance of the assessment from a referential one

The passed data structures must contain ->dimgrades property. The referential assessment is supposed to be close to the average assessment. All dimension grades are supposed to be normalized to the interval 0 - 100. Returned value is rounded to 4 valid decimals to prevent some rounding issues - see the unit test for an example.

Parameters:

stdClass	$assessment	the assessment being measured
stdClass	$referential	assessment
array	$diminfo	of stdclass(->weight ->min ->max ->variance) indexed by dimension id
stdClass	$settings

Returns:

float|null rounded to 4 valid decimals

Reimplemented from workshop_best_evaluation.

Definition at line 51 of file testlib.php.

average_assessment

(

array $

assessments

)

Given a set of a submission's assessments, returns a hypothetical average assessment

The passed structure must be array of assessments objects with ->weight and ->dimgrades properties.

Parameters:

array

$assessments

as prepared by self::prepare_data_from_recordset()

Returns:

null|stdClass

Reimplemented from workshop_best_evaluation.

Definition at line 45 of file testlib.php.

normalize_grades	(	array $	assessments,
		array $	diminfo
	)

Normalizes the dimension grades to the interval 0.00000 - 100.00000

Note: this heavily relies on PHP5 way of handling references in array of stdclasses. Hopefully it will not change again soon.

Parameters:

array

$assessments

of stdclass as returned by {

See also:

self::prepare_data_from_recordset()}

Parameters:

array

$diminfo

of stdclass

Returns:

array of stdclass with the same structure as $assessments

Reimplemented from workshop_best_evaluation.

Definition at line 42 of file testlib.php.

weighted_variance

(

array $

assessments

)

Given a set of a submission's assessments, returns standard deviations of all their dimensions

The passed structure must be array of assessments objects with at least ->weight and ->dimgrades properties. This implementation uses weighted incremental algorithm as suggested in "D. H. D. West (1979). Communications of the ACM, 22, 9, 532-535: Updating Mean and Variance Estimates: An Improved Method" http://en.wikipedia.org/wiki/Algorithms_for_calculating_variance#Weighted_incremental_algorithm

Parameters:

array

$assessments

as prepared by self::prepare_data_from_recordset()

Returns:

null|array indexed by dimension id

Reimplemented from workshop_best_evaluation.

Definition at line 48 of file testlib.php.

---

The documentation for this class was generated from the following file:

• C:/xampp/htdocs/moodle/mod/workshop/eval/best/simpletest/testlib.php