THE MEASUREMENT OF USER INFORMATION SATISFACTION (BLAKE IVES ET.AL) Presented by: IRA GERALDINA

Outline  Introduction  The User Information Satisfaction Construct  Measuring User Information Satisfaction  A Study to Assess a UIS Measure  Improving Instrument Quality  Producing a Short Form  Conclusion

Introduction To develop and validate a measure of UIS that are reviewed and compared Objective

The User Information Satisfaction Construct Cyert & March: An information system must meet the needs of its user It is measurable, but perceptual or subjective measure of system success UIS measures how users view their information system other than the technical quality of the system

The User Information Satisfaction Construct UIS measures ‘s Concepts System effectiveness: felt need, system acceptance, perceived usefulness, feeling about information system, MIS appreciations User involvement in information system development, system usage, and estimate of system value

Measuring User Information satisfaction

Example

A Study to Assess A UIS Measure  Study objectives: (1) replicate Pearson’s findings concerning the validity of the instrument (2) reinforce the validity of the instrument through further tests (3) reduce the length of the overall measure while maintaining reasonable level of reliability and the existing structure of scales (4) develop a standard “short form” instrument

A Study to Assess A UIS Measure  Methodology: - Survey to 800 production managers in U.S. manufacturing organizations - 2 month period for two separate mailings: (1) Pearson’s UIS measure; (2) 4 item measure of information satisfaction/ UIS4 - Final sample: 280 managers (35% response rate)

A Study to Assess A UIS Measure  Results: See Table III & IV - Reliability - Content validity - Predictive validity - Construct validity

Reliability - The amount of error test by Cronbach’s alpha - Reliability scores ranged between 0.82 and scales > Test Retest: UIS4, UIS38, Factor Loaded

Content of Validity  It refers to the representativeness or sampling adequacy  All scales are positive and significant at the level

Predictive Validity  Demonstrated by correlating a measure against other measures of the same construct  A correlation of 0.55 was obtained and significant at the level

Construct Validity  Demonstrated by validating the theory behind the instrument  2 Methods: - Total score >> See column 4 (UIS38) of Table IV, all significant at level - Factor analysis >> See Table III, resulting 5 factors

Factor Analysis

Improving Instrument Quality

Elimination of Scales  Each scale was ranked on the following criteria: (1) reliability (2) content validity (3) construct validity  See Table IV - Triple asterisk were selected for elimination

Reducing the Number of Items per Scale  Holdback sample data was tested  Removing items within a scale that had the lowest correlation with the other items  See Table V, two items were then eliminated

Producing A Short Form  Entails selecting representatives scales that has an acceptable high correlation with the full instrument  The correlation was 0.90 (significant at level)  Contact the authors to have short form questionnaires

Conclusion The instrument presented here is a potential candidate for adoption researchers practitioners

A SHORT FORM MEASURE OF UIS: A PSYCHOMETRIC EVALUATION AND NOTES ON USE BARAUDI & ORLIKOWSKI

The Purpose of Study  To conduct a psychometric evaluation of the short- form of UIS measure are developed by Ives, Olson, and Baroudi  To discuss how the short-form of UIS can be used to (a) detect the presence of user dissatisfaction (b) aid the diagnosis of problem areas

Sample  358 employees from 26 New York area organizations  See Table 1

Sample’s Distribution

The UIS Questionnaire

Validity

Reliability