Search results

Evidence suggests that electronic word-of-mouth (eWOM) plays a highly influential role in decision-making when booking hotel rooms. The number of online sources where consumers can obtain information on hotel ratings provided has grown exponentially. Hence, a number of companies have developed average scores to summarize this information and to make it more easily available to consumers. Furthermore, official star rating schemes are starting to provide these commercially developed average scores to complement the information their schemes offer. The purpose of this paper is to examine the robustness of these systems.

Average scores from different systems, and the scores provided by one rating site were collected for 200 hotels and compared.

Findings suggested important differences in the ratings and assigned descriptive word across websites.

The results imply that the application of average scores by official organizations is not legitimate and identifies a research gap in the area of consumer and star rating standardization.

The paper is of value to the industry and academia related to the examination of rating scales adopted by major online review tourism providers. Evidence of malpractice has been identified and the adoption of this type of scales by official star rating schemes is questioned.

事实证明线上口碑效应（eWOM）在酒店房间预定决策中占有重要位置。消费者能够获取的酒店评定的在线资源已经显著增长。因此, 一些公司已经研发出一套均值评定技术来整合这些信息, 为消费者提供更容易掌握的信息资源。此外, 官方星级评级策略已经开始通过这些商家开发的均值整合技术来完善其星级评定结果。这也就导向了对这些评定系统的科学性提出疑问, 本论文旨在解决这方面的研究。

研究样本包括200家酒店的不同系统均值信息和同一评定网站的评定信息。本论文对其两种来源的评定信息进行了对比。

研究发现不同网站的评级和给与的描述语言差别很大。

研究结果表明采用官方机构指定的均值整合系统并不是合理的, 并指出了消费者和星级评定标准方面的研究有着很大的理论缺口。

本论文对于行业和学术界在大型在线旅游评价网站提供的评定信息方面有很重要的价值。研究事实表明行业错误规范真正存在, 对于采用官方星级评定系统的这种评价存在质疑。

关键词 eWOM、酒店、均值评定、星级评定系统、在线评论

What is the key to library user satisfaction? Can LibQUAL+® help in the quest for delivering a quality library service? The purpose of this paper is to present international research into library customer satisfaction as measured by the LibQUAL+® survey methodology. Commonalities of satisfaction and dissatisfaction have been identified which influence the customers overall view of the library. This knowledge can be used to further increase customer satisfaction through targeting these areas for service improvement.

The LibQUAL+® results from SCONUL Libraries, Utrecht and Leiden Universities were analysed to explore the differences between customers who were very satisfied, and those who were very dissatisfied, with the service. Results from each of the three dimensions of service quality were reviewed separately. The survey results from respondents who had given a high satisfaction mean score to one of the three dimensions were analysed to assess if they had also given high satisfaction mean scores overall. This process was then repeated for those who had given low satisfaction mean scores.

Respondents with high satisfaction mean scores in the Information Control dimension were discovered to have the largest positive scores for the overall average perceived scores, indicating they are the most satisfied customers. When reviewing the surveys with low satisfaction mean scores in the Affect of Service dimension it was discovered that these respondents also had the largest negative scores for the overall average perceived scores, indicating they are the most dissatisfied customers. The findings show that both information resources and customer service affects the overall opinion of the library service for all customer groups.

Good information resources has a positive effect on customers’ opinions of the library just as much as poor service from library staff has a detrimental effect. Any conclusions drawn from these findings should recognise that the research is limited to measuring service quality within the confines of the LibQUAL+® survey methodology. The research has not investigated the reasons for the commonality, nor do these averages say anything about the motivation for each individual respondent to give these scores in the survey.

Statistical analyses confirm that these findings hold for every user group. Therefore, for the library manager seeking to deliver a quality library service it will be important to take both of these factors into account and deliver information not only in a professional, but also in a helpful manner.

Although based on previous research, the extension of the analysis from an institutional level to an international consortia level strengthens the initial research conclusions. The findings, implications, and conclusions are valuable to library managers seeking to improve the customer perceptions of their library service, providing evidence of factors that influence customers’ opinions.

Unlike many other quantitative characteristics used to determine higher education rankings, opinion-based peer assessment scores and the factors that may influence them are not well understood. Using peer scores of US colleges of engineering as reported annually in US News and World Report (USNews) rankings, the purpose of this paper is to provide some insights into peer assessments by statistically identifying factors that influence them.

With highly detailed data, a random parameters linear regression is estimated to statistically identify the factors determining a college of engineering’s average USNews peer assessment score.

The findings show that a wide variety of college- and university-specific attributes influence average peer impressions of a university’s college of engineering including the size of the faculty, the quality of admitted students and the quality of the faculty measured by their citation data and other factors.

The paper demonstrates that average peer assessment scores can be readily and accurately predicted with observable data on the college of engineering and the university as a whole. In addition, the individual parameter estimates from the statistical modeling in this paper provide insights as to how specific college and university attributes can help guide policies to improve an individual college’s average peer assessment scores and its overall ranking.

To enhance and promote quality construction, the Construction Quality Assessment System (CONQUAS) was introduced in Singapore in 1989 to evaluate the quality performance of building contractors using numerical scores. Apart from the CONQUAS system, the ISO 9000 quality management standards were also introduced in the Singapore construction industry in 1991. Is there a relationship between certification to ISO 9000 standards and the achievement of higher construction quality standards as indicated by CONQUAS scores? As several years have passed since the introduction of the CONQUAS system and ISO 9000 in the Singapore construction industry, this paper aims to ascertain if the implementation and certification of quality management systems to ISO 9000 standards in construction firms has helped them to achieve higher construction quality standards through higher CONQUAS scores.

The purpose of this paper is to clarify critical issues underlying the national culture dimensions of Hofstede and GLOBE, demonstrating their irrelevance to international marketing decision‐making.

In‐depth discussion of the theoretical and empirical logic underlying the national culture dimension scales and scores.

Hofstede and GLOBE national culture scores are averages of items that are unrelated and which do not form a valid and reliable scale for the culture dimensions at the level of individuals or organizations. Hence these scores cannot be used to characterize individuals or sub‐groups within countries. The national culture dimension scores are therefore of doubtful use for marketing management that is concerned with individual‐and segment‐level consumer behavior.

Researchers should be cautious in using the Hofstede and GLOBE national culture dimension scores for analysis at the level of individuals and organizations.

Hofstede and GLOBE dimension scores should not be used to infer individual/managerial and group/organizational level behavior and preferences.

The paper follows a recent paper in IMR which was the first to discuss the common misunderstanding of the Hofstede and GLOBE national culture scales and scores, and their misapplication at the level of individuals and organizations by scholars and practitioners. Here we further expand and clarify the issues.

Using the Program for International Student Assessment (PISA) surveys (2003–2015), this chapter explores the relationship between the gender gap in math test scores and computer (digital devices) gaming, as a potential “swimming upstream” factor in the quest to close that gap. Using a decomposition based on a pooled hybrid specification, we attribute two to three points (from 13% to 29%) of the gender math gap to gender differences in the incidence and returns to intense gaming. The comparison of the negative versus positive girl-specific effects found for collaborative games versus single-player games suggest a potential role for gaming network effects.

In the public sector, Training and Experience (T & E) exams assess prior experience and are one of the most often used methods for selecting job applicants. This study uses a KSA approach, where raters judge the quality of job relevant prior experience, not its duration or quantity. It was hypothesized that an additional rater and a consensus meeting between raters would increase reliability and validity.

T & E and supervisory ratings were obtained over a 12-year period for 166 candidates seeking promotion to a budget analyst position. Validity was measured by the correlation between T & E scores and supervisory ratings. Consensus was required only for T & E scores differing by a specific amount (hybrid consensus).

Intraclass reliability was 0.73, 0.84, and 0.95 in the one-rater, two-rater, and hybrid consensus conditions with each coefficient greater than the next (p < 0.05) showing the benefit of multiple raters and consensus for reliability. Validity was significant at 0.21, 0.26, and 0.251 for each rating condition, respectively (two-tail test; p < 0.01). Validity was greater in the two-rater condition than in the one-rater condition (one-tail test; p < 0.05). Consensus did not improve validity beyond that of two raters. For consensus T & Es (n=76), two raters improved validity (one-tail test; p < 0.05), moving from 0.112 to 0.231 but not reliability; consensus improved reliability (two-tail test; p < 0.05) but not validity.

There has been a vacuum in T & E research for close to 20 years. Validity data are difficult to obtain but critical for meta-analysis. T & Es showed validity. Use of two raters improved validity but consensus did not increase the gain.

In two case studies of research units, reference values used to benchmark research performance appeared to show contradictory results: the average citation level in the subfields (FCSm) increased world‐wide, while the citation level of the journals (JCSm) decreased, where concomitant changes were expected. Explanations were sought in: a shift in preference of document types; a change in publication preference for subfields; and changes in journal coverage. Publishing in newly covered journals with a low impact had a negative effect on impact ratios. However, the main factor behind the increase in FCSm was the distribution of articles across the five‐year block periods that were studied. Publication in lower impact journals produced a lagging JCSm. Actual values of JCSm, FCSm, and citations per publication (CPP) values are not very informative either about research performance, or about the development of impact over time in a certain subfield with block indicators. Normalized citation impact indicators are free from such effects and should be consulted primarily in research performance assessments.

This chapter summarizes the quantitative literature on whether intervention outcomes for students with learning disabilities (LD) are influenced by variations in IQ and reading level. The analysis clearly shows that a significant intelligence×reading level interaction emerges in treatment outcomes. Across a broad array of interventions it was found that studies which include samples with reading and IQ scores in the 16th and 25th percentile range (standard scores between 84 and 91) yield significantly higher effect sizes than studies that include samples in same low reading range but with higher IQ scores. An analysis of subsets of this data yield similar findings. Implications for definitions of learning disabilities that include measures of intelligence are discussed.