Assessing Relative Importance Using RSP Scoring to Generate Variable Importance Factor (VIF)

Typically developing, American adolescents in Grade 10 ( n = 38) ranked the relative importance of 14 communication skills in their friends’ communication with them. The skills were considered those associated with figurative language (e.g., appropriate slang use), empathy (e.g., vocal tone comprehension), or discourse management (e.g., appropriate topic selection). Five skills emerged as more important and were primarily considered to be empathy-related communication skills. Four skills, representing a mix of figurative language and discourse management skills, were ranked as lower in importance. The remaining five skills were of intermediate or variable importance and a mix of discourse management and empathy-related communication skills. Results were similar to those of previous investigations with Australian adolescents. Adolescents’ gender did not result in different rankings. Knowing what adolescents opine to be more and less important communication skills for positive peer relationships can help speech-language pathologists decide on intervention objectives to improve adolescents’ peer relationships.

Download Full-text

Adolescents’ Perceptions of the Relative Importance of Selected Communication Skills in Their Positive Peer Relationships

Language Speech and Hearing Services in Schools ◽

10.1044/0161-1461.2603.263 ◽

1995 ◽

Vol 26 (3) ◽

pp. 263-272 ◽

Cited By ~ 15

Author(s):

Felicity Mobbs Henry ◽

Vicki A. Reed ◽

Lindy L. McAllister

Keyword(s):

Peer Relationships ◽

Communication Skills ◽

Variable Importance ◽

Female Adolescents ◽

Relative Importance ◽

High Importance ◽

Male And Female ◽

Intermediate Variable ◽

Tenth Grade ◽

Discourse Management

This study investigated the opinions of tenth grade adolescents with normal achievement concerning the relative importance of 14 communication skills in their positive peer relationships and explored whether or not the adolescents’ gender influenced their opinions about the importance of these skills. Results suggested that two skills were of relatively high importance, four skills were of relatively low importance, and the remaining eight skills were of intermediate or variable importance. Communication skills closely associated with characteristics of empathy and primarily addressee-focused were ranked as either relatively high or intermediate/variable in importance. Those skills related to figurative and metalinguistic language and considered primarily speaker-focused were ranked as relatively low in importance. Communication skills related to discourse management were ranked as being either relatively intermediate/variable or low in importance. No significant differences were found between the overall rankings of the male and female adolescents. Discussion of the findings relates the results to suggestions in the literature concerning communication skills thought to be important for successful adolescent peer relationships, raises directions for further research, and presents implications for intervention.

Download Full-text

MODULAR FEATURE SELECTION USING RELATIVE IMPORTANCE FACTORS

International Journal of Computational Intelligence and Applications ◽

10.1142/s1469026804001021 ◽

2004 ◽

Vol 04 (01) ◽

pp. 57-75 ◽

Cited By ~ 1

Author(s):

SHENG-UEI GUAN ◽

FANGMING ZHU ◽

PENG LI

Keyword(s):

Feature Selection ◽

Classification Accuracy ◽

Feature Space ◽

Data Sets ◽

Relative Importance ◽

Classification Problems ◽

Feature Selection Technique ◽

Importance Factor ◽

Rule Sets ◽

New Feature

Feature selection plays an important role in finding relevant or irrelevant features in classification. Genetic algorithms (GAs) have been used as conventional methods for classifiers to adaptively evolve solutions for classification problems. In this paper, we explore the use of feature selection in modular GA-based classification. We propose a new feature selection technique, Relative Importance Factor (RIF), to find irrelevant features in the feature space of each module. By removing these features, we aim to improve classification accuracy and reduce the dimensionality of classification problems. Benchmark classification data sets are used to evaluate the proposed approaches. The experiment results show that RIF can be used to determine irrelevant features and help achieve higher classification accuracy with the feature space dimension reduced. The complexity of the resulting rule sets is also reduced which means the modular classifiers with irrelevant features removed will be able to classify data with a higher throughput.

Download Full-text