Surveys More on Scaling Essay
Excerpt from Essay :
Scale to Measure Celebrity Endorsers
In this study by Ohanian, a scale used in measuring celebrity endorsers' expertise, trustworthiness as well as attractiveness is developed. Psychometric scale development protocols are followed for testing the data reliability and validity as well. In addition, this study uses two exploratory and confirmatory samples to initiate a 15-item scale measuring the celebrity endorsers' characteristics. This article complies fully with Churchill's recommendations in several fronts as outlined below.
Several sources are researched on to identify words, phrases, and adjectives to use for this research's questionnaire; this results in the development of several adjectives describing personality traits. During the construction of the scale, 182 adjectives are identified of which some were eliminated to about 139 adjectives. Additionally, the 139 descriptors were further trimmed by a group of 38 college students; the researcher believed the words were unfamiliar with respondents, to 104.
For the identification of celebrities for the study, 40 college students named celebrities they could recall in three minutes. The most frequently mentioned names were John McEnroe and Linda Evans who had been involved in advertisements. Celebrities often mentioned, who had not been involved in advertisements, were Tom Selleck and Madonna. Afterwards, a group of 38 students indicated their level of familiarity with each of the celebrities mentioned. In addition, the participants were to name the most and least appropriate products the celebrities sponsored. The results from these samples were used as guides following the inclusion of frequently purchased products used by most individuals in the data collection. This made the final list stand as Linda Evans promoting new perfumes; Madonna, new line of designer jeans; John McEnroe, tennis rackets; and Tom Selleck, new line of men's cologne.
The validation and purification of the data was in two phases; exploratory and confirmatory phases. All questionnaire items were factor analyzed, using principal components analysis then varimax rotation. The initial factor solution for the sample evaluating Madonna resulted in four factors with eigenvalues greater than one. This four-factor solution accounted for 68.4% of the variance. In order to purify the list, items with loadings of 0.3 or greater on more than one of the factors were eliminated and the reduced list factor analyzed again. The result was three factors having eigenvalues greater than one, accounting for 61% of the variance. Additionally, the final items from each factor analysis were tested for reliability by submitting them to item analysis using item-to-total correlations. The items for each subscale were analyzed separately. To obtain a practical size scale, items with the lowest item-to-total correlations were deleted while maintaining an acceptable level of reliability as measured by Cronbach's alpha (Ohanian, 1990). To determine the reliability of subscales for different celebrities and genders, Cronbach's alpha was computed for both male and female respondents for Madonna and John McEnroe. The results indicated a highly reliable scale. Both male and female respondents had equally reliable response patterns, and the total sample for each subscale and celebrity had a reliability coefficient of 0.8 or higher. Using the LISREL methodology to verify the relationship between observable variables and latent constructs, two confirmatory factor-analysis models (Linda and Tom Selleck) were tested separately. The x2 statistic was non-significant for each model, indicating an adequate fit of the confirmatory model to the data. (X2SELLECK= 99.60, df = 87, p = 0.168; X2EVANS =109.71, df =87, p = 0.051). The root mean square residual was 0.048 and 0.046 for Selleck and Evans respectively.
Madonna Sample (n = 249)
John McEnroe (n = 237)
Compatible with the Product
Informed about the Product
Familiar with the
scale does scale discriminate within and between responders. It gives a fair result from the data collected and clear analysis as recommended by Churchill in his publication (Churchill, 1979). Just like any other research, this study has challenges it faces. First, being a quantitative analysis, it only establishes the reliability and validity of the scale rather than discovering their existence. In line with this, the existing scale can undergo modification. Besides, from the large amount of money spent on celebrity advertising, advertisers should use the scale as an integral part of their effectiveness testing and tracking.
The RELQUAL Scale
In this publication, a RELQUAL scale is developed to assess the degree of relationship quality between exporting firms and importers. This article too is to a greater degree compliant to the specifications issued by Churchill. The analysis conducted in this publication explains whether reliability, data convergence as well as validity are attainable from the instrument used for data collection. This study uses four multi-item scales in conducting out the research. This study sample involved 1,564 British enterprises of which, 111 replies were returned in 2002; a response rate of 7%.
Churchill (1979) notes that for high reliability and low measurement errors, a multi-item scales instead of single-item scales should be used. This research uses Confirmatory Factor Analysis with each item restricted to load on its pre-specified factor, with the four first order factors allowed to correlate freely. From this study, the chi-square for the collected data is significant (?2=125.97, df = 71, and p<.05). Moreover, the Comparative Fit Index, Incremental Fit Index, and Tucker-Lewis Fit Index of this model are 0.92, 0.92, and 0.90, respectively. This shows the final structural model is good in reproducing the population covariance structure, and there exists an acceptable discrepancy between the covariance matrices.
The items used in this scale are valid. Therefore, all the construct inter-correlations are significantly different from 1, while the square of their inter-correlation is less than the average variance (Lages, Lages, & Lages, 2005). In this study, discriminant validity is tested by including in the model an established construct (?=.83; vc (n)=.50; =.83). Discriminant validity is revealed by non-significant correlations among the four first-order constructs and the new construct.
In addition, an EXPERF scale is developed to be applied at the export venture level and all of the three dimensions are shown to be reliable and valid: financial export performance (?=.80; vc (n)=.61; =.82), strategic export performance (?=.90; vc (n)=.77; =.91), and satisfaction with export venture (?=.90; vc (n) =.75; =.90).
Financial Export Performance
Strategic Export Performance
Satisfaction with Export Venture
Amount of information sharing in the relationship
Communication quality in the relationship
Long-term relationship orientation
Satisfaction with the relationship
The numbers outside the brackets are the coefficients while the ones bracketed are the t-values. From this table, *p<