Data quality control for survey instrument of office investors in rationalising green office building investment in Kuala Lumpur by the application of Rasch analysis

The purpose of this paper is to attempt to verify the quality of the survey instrument of office investors in rationalising green office building investment in the city of Kuala Lumpur using the Rasch measurement method. It investigates whether the quality of the data obtained from the survey is statistically acceptable and aims to ensure that the scales used are based on the same measurement model and fit with the Rasch model.

In achieving this objective, a questionnaire survey was developed consisting of six sections. Some 394 questionnaires were distributed, and in total, 106 responses were received from office investors who own and lease office buildings in the Federal Territory of Kuala Lumpur. The data obtained from this survey were entered into Rasch software, and the analysis aims to consider three main parameters, specifically: point measure correlation 0.32 < x < 0.8; outfit mean squared 0.5 < y < 1.5; and outfit z-standard −2 < z < 2. It also provides a separable or independent measurement instrument for the parameters of the research object.

The analysis performed using the Rasch model confirmed that all items in the questionnaire construct were statistically reliable and valid. The Rasch analysis consists of, namely, the summary statistics; item unidimensionality to provide interval measures of item endorsements and fit statistics on persons involved; and items for further investigation. Unidimensionality, as a pre-requisite to the Rasch analysis, provides statistics on whether the items are on the same latent variable intended by the instrument. The results were also supported by Cronbach’s alpha at 0.91 which showed excellent reliability for person data. The results showed that the summary statistics, unidimensionality and item fit analysis were excellent.

This paper introduces the application of the Rasch analysis as a provision for a new dimension and technique in examining data reliability and validity of the instrument.