A research
Children who are poor in school performance often lack the cognitive abilities like listening accuracy, phonological awareness, working memory, comprehension etc. Listening accuracy is the ability to discriminate acoustic sweeps. Phonological awareness is ability to distinguish parts of speech, such as syllables and phonemes presented auditorily. Working memory implies a system for the temporary holding and manipulation of information during the performance of a range of cognitive tasks.
Fast ForWord Language Gateway Edition product software of Scientific Learning Corporation aims to improve above abilities through it's web based 7 training exercises. These are Block commander (BC), Old MacDonald’s Flying Farm (OM), Phoneme Identification (PI), Phonic Match (PM), Phonic Words (PW), Language Comprehension Builder (LCB), Block Commander (BC).
Aim of this study was to assess criterion related validity of the said product following request from the Scientific learning corporation, USA. Here, validity refers to the extent of content completion of above 7 exercises at the base, middle and terminal level of training.
Initially, Step One foundation for Child and Youth Welfare trained 25 students of St. Mary's Orphanage & Day school, Kolkata with Fast ForWord Language Gateway Edition product. Training response data were used in this analysis. Before the training auditory discrimination ability was assessed by GFW test.
Base level analysis:
As expected 72% of the total trainees were limited within 15% completion irrespective of exercise wise differences. This suggests three things - (i)the product is useful for the students with age range from 9 to 11 years; (ii)difficulty level of stimuli for each exercise is arranged in such a fashion so that no one can achieve the desired outcome at the initial level; (iii)good base level validity of the product.
Middle level
Middle level refers to the period between initial and terminal level of training. The analysis represents changes in cognitive field of trainees during the training. Middle level training performance data analysis is very complex as results may be affected by two types of errors - (a) criteria contamination and (b)confounding effect. Any training performance is affected by one's initial level of competency. This should be properly accounted in training analysis otherwise there will be high possibility for criteria contamination. Like criteria contamination, training is affected by confounding variable. A confounding variable (also confounding factor, lurking variable, a confound, or confounder) is an extraneous variable in a statistical model that correlates (positively or negatively) with both the dependent variable and the independent variable. In this analysis, confounding variable is auditory discrimination capacity of the trainees as most of the exercises are based on auditory stimuli. Therefore, to evaluate the data, three specific questions are addressed : (a)what is happening within cognitive field of the trainees; (b)Do the students with high base line competency show much proficiency than their counterparts ? (c) Do the students with high auditory discrimination ability show much proficiency than the students with relatively low ability ?
For the first question, box plot analysis was used. It revealed usual plateau stage before achieving the target. of different exercises. Plateau was high in case of BC exercise. Trainees experienced much difficulty in case of two phonic based exercises, namely, PI and PW.
For the second question, total trainees were classified into two groups based on their initial level of performance. Next mean proficiency levels of both groups across each exercise was compared. It is noted that at the initial stage, mean
difference was high; gradually it was reduced. However, this reduction was high in case of the PM exercise suggesting proficiency improved with training in the low ability group. In the case of the PW exercise, the difference was high during later sessions suggesting difficulties experienced by the students of low ability group as they developed proficiency.
To answer the third question, total trainees were classified into two groups- (a)good auditory discrimination and (b)poor auditory discrimination. Like above mean differences in proficiency between the groups was estimated. It is noted that trainees with good auditory discrimination was able to complete more contents in CS, OM,PM,PW and BC exercises than the poor auditory discrimination group.
Terminal Level
Terminal level success was measured in two ways: as 100% proficiency and more than 93% proficiency as all the trainees could not achieve 100% proficiency in all the exercises. More number of trainees achieved 100% proficiency in PM, OM,and CS exercises. When 93% proficiency was accounted, besides the above three, more number of trainees showed success in BC,LCB exercises.
To summarize, by the end of the sessions, most trainees reached the desired outcome in at least half of the exercises suggesting usefulness of the product. The analysis reveals differential pattern of criterion validity of different exercises of the product. It is noted that PI and PW exercises possessed poor criterion related validity. This two exercises needed more restructuring in considering Indian accent and pronunciation.
Fast ForWord Language Gateway Edition product software of Scientific Learning Corporation aims to improve above abilities through it's web based 7 training exercises. These are Block commander (BC), Old MacDonald’s Flying Farm (OM), Phoneme Identification (PI), Phonic Match (PM), Phonic Words (PW), Language Comprehension Builder (LCB), Block Commander (BC).
Aim of this study was to assess criterion related validity of the said product following request from the Scientific learning corporation, USA. Here, validity refers to the extent of content completion of above 7 exercises at the base, middle and terminal level of training.
Initially, Step One foundation for Child and Youth Welfare trained 25 students of St. Mary's Orphanage & Day school, Kolkata with Fast ForWord Language Gateway Edition product. Training response data were used in this analysis. Before the training auditory discrimination ability was assessed by GFW test.
Base level analysis:
As expected 72% of the total trainees were limited within 15% completion irrespective of exercise wise differences. This suggests three things - (i)the product is useful for the students with age range from 9 to 11 years; (ii)difficulty level of stimuli for each exercise is arranged in such a fashion so that no one can achieve the desired outcome at the initial level; (iii)good base level validity of the product.
Middle level
Middle level refers to the period between initial and terminal level of training. The analysis represents changes in cognitive field of trainees during the training. Middle level training performance data analysis is very complex as results may be affected by two types of errors - (a) criteria contamination and (b)confounding effect. Any training performance is affected by one's initial level of competency. This should be properly accounted in training analysis otherwise there will be high possibility for criteria contamination. Like criteria contamination, training is affected by confounding variable. A confounding variable (also confounding factor, lurking variable, a confound, or confounder) is an extraneous variable in a statistical model that correlates (positively or negatively) with both the dependent variable and the independent variable. In this analysis, confounding variable is auditory discrimination capacity of the trainees as most of the exercises are based on auditory stimuli. Therefore, to evaluate the data, three specific questions are addressed : (a)what is happening within cognitive field of the trainees; (b)Do the students with high base line competency show much proficiency than their counterparts ? (c) Do the students with high auditory discrimination ability show much proficiency than the students with relatively low ability ?
For the first question, box plot analysis was used. It revealed usual plateau stage before achieving the target. of different exercises. Plateau was high in case of BC exercise. Trainees experienced much difficulty in case of two phonic based exercises, namely, PI and PW.
For the second question, total trainees were classified into two groups based on their initial level of performance. Next mean proficiency levels of both groups across each exercise was compared. It is noted that at the initial stage, mean
difference was high; gradually it was reduced. However, this reduction was high in case of the PM exercise suggesting proficiency improved with training in the low ability group. In the case of the PW exercise, the difference was high during later sessions suggesting difficulties experienced by the students of low ability group as they developed proficiency.
To answer the third question, total trainees were classified into two groups- (a)good auditory discrimination and (b)poor auditory discrimination. Like above mean differences in proficiency between the groups was estimated. It is noted that trainees with good auditory discrimination was able to complete more contents in CS, OM,PM,PW and BC exercises than the poor auditory discrimination group.
Terminal Level
Terminal level success was measured in two ways: as 100% proficiency and more than 93% proficiency as all the trainees could not achieve 100% proficiency in all the exercises. More number of trainees achieved 100% proficiency in PM, OM,and CS exercises. When 93% proficiency was accounted, besides the above three, more number of trainees showed success in BC,LCB exercises.
To summarize, by the end of the sessions, most trainees reached the desired outcome in at least half of the exercises suggesting usefulness of the product. The analysis reveals differential pattern of criterion validity of different exercises of the product. It is noted that PI and PW exercises possessed poor criterion related validity. This two exercises needed more restructuring in considering Indian accent and pronunciation.
Comments