Language is a tool for communication and it determines the competence of a person in the communicative context. It is referred to [Carroll and Hall, 1985, p.167] interpersonal communication rather than to internal thinking, because it is such interaction which forms the focus for teaching and testing. Nonverbal codes of communication, by gestures, pictures or symbols are peripherally taken into account by us in judging communication.
Language Laboratory Tests are designed to administer through language laboratory. The responses of the examinees usually recorded and weighted. Various aspects of listening comprehension, sound discrimination, speaking, reading (speed), writing (dictation pattern), etc., are assessed by administering these tests. They acquire importance in view of their providing feedback to the learners especially in L2 context.
Language Test is one that is designed and administered for assessing the students in their acquiring different language learning skills, such as Receptive skills (listening and reading)and Productive skills (speaking and writing). These tests are composed by using phonetic, lexical, syntactic and semantic contents of the language. {Also see, Proficiency Test}.
Latent Trait (Measurement) Theory is also known as Item Response Theory. This is of item analysis. Primarily it refers to three families of analytical procedures, viz., the one parameter, two parameter, and three parameter logistic models. These models offer a systematic procedure to quantify the probability or improbability of individual item and person response patterns, besides offering improved ways for the estimation of item difficulty and person ability. Scale of 'person ability' and 'item difficulty' is taken as the first parameter; continuous estimate of discriminability is the second parameter and the 'index of guessing' is the third parameter. Depending on the number of parameters incorporated, these models differ in complexity. Following are the advantages of these models as listed out by Henning [1987, pp.108-115]:
n Sample-free item calibration.
n Test-free person management.
n Multiple reliability estimation.
n Identification of guessers.
n Potential case of administration and scoring.
n Economy of items.
n Reconciliation of norm-referenced and criterion-referenced testing.
n Test equating facility.
n Test tailoring facility.
n Item banking facility.
n The study of item and test bias.
n Elimination of boundary effects in programme evaluation.
n Item and person fit validity measures.
n Score reporting facility.These advantages are universally available through the Rasch (one-parameter) model, but may or may not be available in the other two models.
Learner Behaviour refers to any visible activity (scholastic and non-scholastic) displayed by a learner.
Likert Scale is a five-point rating scale (points ranging from strongly agree to strongly disagree) used to determine comparative magnitudes of some attitudes or opinions founding the respondent.
Length Cues are the cues that are to be avoided while preparing the multiple choice items. Mostly the longest and most explicit option available will obviously provide a clue to identify the answer. Hence, all the options provided are expected to be more or less equal in length.
Leplokurtik [See, kurtosis].
Linear Regression is one wherein the association between two variables is described by a straight line. If the association is not linear, the variables ca be transformed in such a way that the association becomes linear for applying the technique of linear for applying the technique of linear regression for estimating the coefficients.
Linear Transformation is a conversion process and is said to have taken place when a constant or set of constant is arithmetically applied to a distribution of scores for the purpose of changing the form of the distribution. For instance, conversion of raw scores into percentage scores.
Linear Transformation is a conversion process and is said to have taken place when a constant or set of constants is arithmetically applied to a distribution. For instance, conversion of raw scores into percentage scores.
Line of Best Fit is another term for the regression line (used in regression analysis) [Fig. in appendix].
Listening Test is meant to evaluate the ability of the students to listen and comprehend various kinds of speeches, lectures delivered in a classroom, radio, TV, cinema, etc., both in the formal and informal situations.
Logical Validity is an estimate of content validity based on the comparison of the behaviour demanded by the test with the behaviour that belongs to the variable to be measured. It compares [English and Ava] the behaviours elicited by the test items to see whether there is correspondence. It is also sometimes referred to as 'rational validity'. [Syn.: Construct Validity, a priori Validity].
Logit is a unit of measurement of an interval latent trait scale.
Long Answer Questions are otherwise known as essay questions and for which the responses will usually be lengthy. They are used to test intellectual higher order abilities, such as comprehension, recall, application, analysis, synthesis and evaluation. They can be of two types :
1) Restricted Response.
2) Unrestricted Response(Also see, Essay Questions).
Lower Order Abilities - the first three levels of the cognitive behaviour, viz., knowledge, comprehension and application are often referred to as the lower order abilities. They are invariably content oriented, content dependent and are controlled by the nature, depth and quantity of content. [Also see, Scholastic Ability].