A Chinese version of Kogan's Attitude toward Older People Scale: Reliability and validity assessment

1. Introduction 

According to the Taiwan census, the elderly population will approach 20% within the next 20 years. This enhances the need for skilled and experienced healthcare providers, in many health care settings, who can address and meet the needs of older people. Physicians and nurses have a pivotal role as providers of care for older people and are uniquely positioned to influence the quality of their care (Stevens and Herbert, 1997). Therefore, healthcare providers in Taiwan need to increase their direct and indirect work with the elderly. However, previous studies have indicated that healthcare providers do not enjoy taking care of the elderly due to ignorance and poor attitude (Palmore, 1999). As the population grows older, attitudes towards the elderly require evaluating. However, no such exploration on health-related students’ attitudes towards older people has yet been carried out in Taiwan. In order to provide better care to the elderly population, exploring the students attitudes toward older people is a vital concern. The Kogan's Attitude toward Older People (KAOP) scale has been found to possess high reliability and validity values in several international research studies; however, there has not been any verification of the Chinese version. This was the need that this study sought to fill. Consequently, it could be used as a measurement of KAOP for both health-related students and clinicians.

2. Methods 

3. Results 

3.1. Demographic data 

275 students returned the KAOP. Demographic data of students are summarized in Table 1. Medical students were in the majority (n=145), as were sophomores. The sample mean score was 144.30 (S.D.=17.89), with approximately half of the subjects (50.1%) scoring above the mean. As shown in Table 1, females had higher scores (more positive attitudes), nursing students scored higher than medical students, and sophomores rated higher than seniors.

Table 1. Summary of major, year of study, gender, and mean score

		n	M	S.D.
	Major (n=275)
	Medicine	191	138.77	15.04
	Nursing	84	156.86	17.56
	Year of study (n=133)
	Sophomore	196	146.01	18.59
	Senior	79	140.04	15.31
	Gender
	Male	140	140.47	15.93
	Female	135	148.27	18.97

3.2. Validity 

3.2.1. Content validity 

The content validity index (CVI) was used to determine item validity. Panel experts (5 general geriatricians, 1 faculty member from the Department of Family Medicine, 2 gerontological experts and 1 school administrator) were asked to rate each item of the Chinese version of the KAOP based on relevance, clarity, and simplicity as 1 (not relevant), 2 (somewhat relevant), 3 (relevant), or 4 (very relevant). A CVI was computed using the proportion of experts who were in agreement about item relevance. The average CVI was .92 in the final version indicating adequate content validity (>.80; Polit and Beck, 2004).

3.2.2. Construct validity 

Construct validity was supported in the factor analysis. The numbers of factors were determined by eigenvalues (≧1), Kaiser–Meyer–Olkin (KMO), scree plots, loadings of over .40 (Stevens, 1996), and explainable percentages of variance. The KMO was 0.91 indicating sampling adequacy (>.50; Kaiser, 1974). Bartlett's test of sphericity was statistically significant. Two factors were extracted and explained 54.7% of the variance (>70%; Stevens, 1996).

All 34 items demonstrated moderate to strong loading (>.40; See Table 2). Factor 1, labeled Prejudice, consisted of 17 negative items, explaining 33.6% of the variance. Factor 2, labeled Appreciation, consisted of only positive items, explaining 21.1% of the variance. There are some similarities with other studies (Hweidi and Obeisat, 2006, Lambrinou et al., 2005), such as the items expressing Prejudice and Appreciation.

Table 2. Factor loadings after varimax rotation for the KAOP scale

	Item	Item content	Factor 1	Factor 2
			Prejudice	Appreciation
	15N	The elderly are irritable, grouchy and unpleasant	.88	.21
	12N	The elderly have a negative influence on a neighborhood	.82	.19
	10N	The elderly are always prying into the affairs of others	.80	.02
	14N	The elderly are untidy	.76	.04
	16N	The elderly complain about the young	.72	.17
	9N	The elderly bore others with their stories	.71	.26
	5N	The elderly have shabby home	.68	.26
	11N	The elderly have irritating faults	.66	.05
	8N	The elderly make others feel ill at ease	.64	.01
	13N	The elderly are much alike	.63	.17
	3N	The elderly are unable to change	.59	.05
	17N	The elderly have excessive demands for love	.57	.09
	7N	The elderly have too much influence in society	.54	.27
	4N	The elderly quit work when they become pensioners	.52	.09
	6N	Wisdom does not come with advancing age	.49	.03
	1N	The elderly should live in special residences	.46	.15
	2N	The elderly are different	.41	.10
	10P	The elderly mind their own business	.24	.75
	5P	The elderly have clean, attractive homes	.03	.70
	7P	The elderly should have more power in society	.03	.69
	8P	The elderly are relaxing to be with	.06	.68
	15P	The elderly are cheerful, agreeable and good-humored	.15	.66
	6P	The elderly grow wiser with advancing age	.27	.65
	14P	The elderly are clean and neat	.21	.65
	4P	The elderly prefer to work as long as they can	.28	.60
	9P	It is nice when the elderly speak about their past	.15	.58
	16P	The elderly seldom complain about the young	.08	.51
	3P	The elderly are capable of new adjustment	.23	.50
	1P	The elderly should live in special residences	.09	.49
	12P	Neighborhoods are nice when integrated with the elderly	.20	.48
	17P	The elderly need no more love than others	.01	.46
	13P	The elderly are different from one another	.07	.45
	11P	The elderly have the same faults as the young	.25	.43
	2P	The elderly are no different from anyone else	.02	.43

3.3. Reliability 

3.3.1. Internal consistency reliability 

Cronbach's alpha was 0.82 for the total scale, 0.83 for Prejudice, and 0.81 for Appreciation which indicate good internal consistency reliability.

3.3.2. Stability reliability 

The Pearson-correlation coefficients for Prejudice and Appreciation were 0.86 and 0.91. Stability reliability was adequate in this study.

3.3.3. Means and standard deviation 

The means and standard deviation of each item are presented in Table 3. Thirteen out of 17 negative items had higher mean scores than positive items.

Table 3. Means, standard deviations, spearman's rank correlations, and item-total correlation (Rs) for KAOP items (N=275)

	Item	Item content	M	S.D.	Rs
	1N	The elderly should live in special residences	4.04	1.84	.31*
	1P	The elderly should live integrated with the young	5.01	1.15	.42*
	2N	The elderly are different	3.12	1.56	.30*
	2P	The elderly are no different from anyone else	4.11	1.45	.31*
	3N	The elderly are unable to change	5.29	1.19	.51*
	3P	The elderly are capable of new adjustment	5.15	1.17	.47*
	4N	The elderly quit work when they become pensioners	5.01	1.28	.32*
	4P	The elderly prefer to work as long as they can	5.21	1.13	.56*
	5N	The elderly have shabby homes	6.07	1.06	.49*
	5P	The elderly have clean, attractive homes	4.56	1.36	.40*
	6N	Wisdom does not come with advancing age	3.42	1.55	.34*
	6P	The elderly grow wiser with advancing age	3.99	1.51	.45*
	7N	The elderly have too much influence in society	3.20	1.39	.34*
	7P	The elderly should have more power in society	4.57	1.45	.54*
	8N	The elderly make others feel ill at ease	5.20	1.37	.54*
	8P	The elderly are relaxing to be with	4.43	1.43	.60*
	9N	The elderly bore others with their stories	5.01	1.56	.50*
	9P	It is nice when the elderly speak about their past	5.46	1.12	.50*
	10N	The elderly are always prying into the affairs of others	4.89	1.41	.46*
	10P	The elderly mind their own business	4.68	1.41	.62*
	11N	The elderly have irritating faults	4.64	1.53	.59*
	11P	The elderly have the same faults as the young	5.40	0.97	.33*
	12N	The elderly have a negative influence on a neighborhood	5.24	1.35	.54*
	12P	Neighborhoods are nice when integrated with the elderly	5.70	1.24	.52*
	13N	The elderly are much alike	4.50	1.64	.45*
	13P	The elderly are different from one another	5.45	1.36	.44*
	14N	The elderly are untidy	5.76	0.94	.55*
	14P	The elderly are clean and neat	4.71	1.34	.62*
	15N	The elderly are irritable, grouchy and unpleasant	5.48	1.16	.58*
	15P	The elderly are cheerful, agreeable and good-humored	4.65	1.35	.44*
	16N	The elderly complain about the young	5.21	1.37	.48*
	16P	The elderly seldom complain about the young	3.25	1.22	.38*
	17N	The elderly have excessive demands for love	2.25	1.33	.34*
	17P	The elderly need no more love than others	2.17	1.15	.30*

* p<.001.

4. Discussion 

The Chinese version of the KAOP indicates good content validity because the CVI was .92. Construct validity was supported in the factor analysis. The factor analysis of the KAOP shows two factors. Two factors were explained 54.7% of the variance. The 54.7% of the variance seems not high but acceptable when compared with Greek version of KAOP including 6 factors with explaining the 41.5% of the variance (Lambrinou et al., 2005). There are some similarities with other studies, such as the items expressing “Prejudice” and “Appreciation” (Hweidi and Obeisat, 2006). Additionally, the KAOP shows good internal consistency reliability as well as stability.

Results indicate that the KAOP− scale is stronger and has higher scores than the KAOP+ scale, which is also in line with previous studies (Kogan, 1961). However, Chinese students who took part in this study reported slightly more positive attitudes when compared with Jordanian students (Hweidi and Obeisat, 2006).

As in most of developed countries, Taiwan is facing the dilemma of a rapidly growing elderly population and there is a perceived need for greater gerontological education for health care professionals. Several universities have recently started providing elective certification programs on gerontology. In order to provide useful and sufficient information to increase students’ knowledge and attitudes, to develop a reliable and valid instrument is essential and helpful for faculty as well as the program director to pinpoint students’ strengths and weaknesses. Educators in many academic programs, including medical and nursing curricula, must better prepare students for the booming elderly population.

5. Conclusions 

Psychometric analysis of the Chinese version of KAOP scale indicates high reliability (internal consistency and stability) and good content and construct validity. Results also reveal two different factors underlying student attitudes toward older people—one related to prejudice and the other associated with appreciation. Based on these findings, it is recommended that the Chinese version of the KAOP be incorporated or adapted by health-related professions as one of the tools used to assess the students’ and/or health care providers’ attitudes toward the elderly.

The KAOP− scale has higher scores than the KAOP+ scale in this study sample, which indicates more emphasis on gerontological curricula and training in undergraduate programs, is strongly needed. Hopefully, the medical and nursing programs could help in preparing health-related professionals capable of meeting the needs of older people by using the Chinese version of the KAOP scale.

Conflict of interest 

None declared.