The etiology of mental disorders: Public endorsement of traditional Chinese beliefs in Macau

Abstract

Background:

Public beliefs concerning the causes of mental disorders are important in their effective treatment. The relative importance of public beliefs related to the etiology of mental disorders among Chinese populations and their relationship to other attitudes to mental disorders are poorly understood.

Aims:

To investigate the endorsement of beliefs related to the etiology of mental disorders in Macau, in particular, traditional Chinese beliefs, and to explore their relationship to general attitudes towards mental disorders.

Methods:

A survey of 361 members of the public measured attitudes towards 32 possible causes of mental disorders as well as contact with and social distance from individuals with a mental disorder.

Results:

The results indicated that traditional Chinese beliefs were ranked with relatively low importance compared with psychosocial causes. Traditional beliefs related to two underlying factors and were significantly higher in participants with lower education levels. There was a significant negative correlation between endorsement of traditional beliefs and contact with individuals with a mental disorder and a significant positive correlation between endorsement of traditional beliefs and desired social distance from individuals with a mental disorder.

Conclusion:

Factors underlying traditional Chinese beliefs concerning the causes of mental disorders are associated with more negative attitudes towards individuals with a mental disorder, with such beliefs more strongly endorsed by those with relatively lower education levels.

Keywords

Mental disorder etiology lay theories Chinese

Introduction

Public beliefs and attitudes (‘lay theories’) concerning the nature, causes and treatment of mental illness are an area of increasing research interest. The study of lay theories is important as there is evidence that they influence the likelihood of seeking treatment, the acceptance of treatment and the satisfaction with treatment (Atkinson, Worthington, Dana & Good, 1991; Callan & Littlewood, 1998; Chen & Mak, 2008; Kelly, Mamon & Scott, 1990). Lay theories have been studied among the general public (Angermeyer & Matschinger, 1996; Furnham & Bower, 1992), including Chinese populations (Furnham & Chan, 2004; Luk & Bond, 1992; Mathews, 2011). Etiology beliefs have also been studied in clinical populations (Holzinger, Loffler, Muller, Priebe & Angermeyer, 2002; McCabe & Priebe, 2004; Saravanan et al., 2007). The present study focuses on ‘etiology beliefs’, which refers to beliefs held by individuals (in either clinical or non-clinical populations) in relation to the cause of mental illness. Such etiology beliefs are an important component of both lay theories related to mental illness and the explanatory models of individuals in clinical populations (Kleinman, 1980). The present study investigated the frequency with which different etiology beliefs relating to mental illness are endorsed by the public in Macau, focusing on etiology beliefs rooted in traditional Chinese culture. The relationship between the endorsement of traditional beliefs and more general attitudes towards mental illness (a desire for ‘social distance’ from the mentally ill) was also explored.

Luk and Bond (1992) reported one of the earliest studies of etiology beliefs in a Chinese population. They conducted a survey of 222 Chinese people in Hong Kong (the majority of whom (84%) were below 30 years of age, with about half attending university). Ten versions of the questionnaire were used, with each version relating to one specific psychological problem/disorder. The 10 disorders were agoraphobia, anorexia nervosa, compulsive gambling and schizophrenia (as in Furnham & Henley, 1988), and also six problems identified by interviews with 10 individuals from a university in Hong Kong. These six problems included two culturally specific disorders (neurasthenia and shen-kwei (a culture-specific psychosexual syndrome) and four problems more social in nature (e.g. social apathy). Participants completed only one of the 10 possible questionnaires and, after reading a description of the problem, they rated the importance of 26 possible causes (e.g. ‘Whether the problem is caused by the person’s genetic predisposition’) on a 10-point scale (0 = completely unimportant, 9 = extremely important). A factor analysis of responses indicated two underlying factors (which together accounted for 32.7% of the total variance). Luk and Bond (1992: 156) labelled these factors as ‘environmental/hereditary’ and ‘social-personal’ and proposed that this was evidence for Hong Kong people employing an ‘interactionist model’ concerning the causes of psychological disorders (a combination of external/environmental causes together with internal/physical weaknesses that increased susceptibility).

Furnham and Chan (2004) compared British and Hong Kong Chinese young people’s beliefs concerning the causes and cure of schizophrenia. A total of 339 participants (176 in Hong Kong and 163 in South East England) completed a questionnaire of 60 questions. These included 16 statements about the cause of schizophrenia where participants were asked to indicate the extent to which they disagreed/agreed with each statement. These included genetic (e.g. ‘Schizophrenia is caused by having blood relatives who are schizophrenic’), biological (e.g. ‘Schizophrenia is caused by having a low birth weight’) and psychosocial (e.g. ‘Traumatic experiences in early childhood can cause schizophrenia’) factors. In addition to these statements explicitly related to the cause of schizophrenia, the questionnaire also included 21 statements concerning general beliefs about schizophrenia, three of which related to ‘superstitious/supernatural’ beliefs (e.g. ‘When patients report that they have delusions, what they see or hear are ghosts’). The researchers hypothesized that such superstitious beliefs would be endorsed more strongly by the Hong Kong Chinese participants. However, the results did not support this and no difference was found. Furnham and Chan (2004) suggested that the failure to identify higher levels of superstitious beliefs in the Hong Kong Chinese sample may have been due to the young and educated sample used in the study (as in the study of Luk and Bond (1992)). Therefore, superstitious beliefs may be more common in older age groups and in those with less formal education.

Using the same list of causes as Luk and Bond (1992), Chen and Mak (2008) conducted a study comparing the etiology beliefs of undergraduate students from the USA, Hong Kong and mainland China and related them to their likelihood to seek help for a psychological problem. Mean scores for the environmental/hereditary and social-personal factors indicated variation in attitudes between the different groups of participants. Environmental/hereditary causes were considered to be more important among ethnically Chinese participants (M = 4.36, 5.59 and 5.19 for American Chinese, Hong Kong Chinese and mainland Chinese, respectively) compared with European Americans (M = 3.68). However, social-personal causes were also considered to be more important among ethnically Chinese participants (M = 3.29, 3.84 and 4.10 for American Chinese, Hong Kong Chinese and mainland Chinese, respectively) compared with European Americans (M = 2.86). The results of a path analysis indicated that help-seeking likelihood was positively related to perceiving psychological problems as having environmental/hereditary causes. In contrast, help-seeking likelihood was negatively related to perceiving psychological problems as having social-personal causes. Chen and Mak (2008) suggested that this was because individuals would not feel responsible for their illness if they perceived it as being due to environmental/hereditary causes; therefore seeking help posed less of a threat to their ‘self-worth’.

As well as studies of public/lay beliefs related to the etiology of mental illness in Chinese populations, etiology beliefs have also been explored among clinical populations. For example, Phillips, Li, Stroup and Xin (2000) developed the Causal Models Questionnaire for Schizophrenia, which included a wide range of possible ‘folk’ beliefs concerning etiology. A total of 45 possible causes grouped into six classes/categories were explored among 245 family members at the time of admission for 135 patients (respondents endorsed an average of 2.5 causes). A weighted measure of the importance of reported causes indicated that the most important was ‘personality problems’, followed by ‘stress’, ‘conflict in non-family relationships’, ‘too much thinking’ and ‘problems with marital engagement’. Of the 45 possible causes, nine were not reported by any of the respondents. For the six classes of causes, spiritual causes accounted for less than 12% of attributed causes, with ‘spirit possession’ and ‘fate’ being the two spiritual causes ranked as the most important in this class (ranked 11th and 7th, respectively, among the 36 cited causes). Overall, the results of Phillips et al. (2000) indicate that a wide range of folk beliefs about schizophrenia are present in China, and that spiritual beliefs, although less important than social and psychological factors, are still perceived as a significant cause of schizophrenia.

More recently, Mathews (2011) studied the relationship between religious beliefs and beliefs about the causes of mental disorders among predominantly Chinese (85%) students of several higher education institutions in Singapore. Participants indicated their level of agreement to a wide range of statements about the causes of mental problems. Five underlying factors emerged, two of which were supernaturally oriented (one related to fate/karma, the other to sin/curses). Further analysis indicated that the endorsement of supernaturally based causes was associated with higher levels of religiosity and less exposure to the formal study of psychology.

In Asian populations, underutilization of mental health services is an issue (Abe-Kim et al., 2007; Le Meyer, Zane, Cho & Takeuchi, 2009). However, this does not appear to be due to lesser need; in fact Asians can experience more symptoms in comparison with other groups (Norton, 2007; Young, Fang & Zisook, 2010). In a recent large-scale study of mental health service utilization in Canada, the percentage of Chinese respondents who had contacted mental health professionals over a 12-month period was significantly lower compared with non-Chinese respondents (Chen, Kazanjian & Wong, 2009). This discrepancy could not be accounted for by language barriers; instead it was concluded that culture was the major factor explaining differences between Chinese and non-Chinese Canadians’ utilization of mental health services. Therefore, studies of the beliefs and attitudes of Chinese populations will promote an understanding of the cultural beliefs related to psychopathology and inform strategies to promote mental health within these communities (Chu & Sue, 2011).

The present study investigated the level of endorsement of different beliefs relating to the cause of mental disorders among the adult Chinese population of Macau, a former Portuguese colony, now a special administrative region of China. Although Macau has a greater degree of cultural variation than mainland China due to its exposure to western cultural influences (Gunn, 1996), its ethnic make-up and culture is still predominately Chinese (95% of the population are Chinese, with over 45% having been born in mainland China (MSCS, 2007)). A questionnaire was used to measure the extent to which participants agreed or disagreed with a range of statements relating to possible causes of a psychological/mental illness (e.g. ‘stress can cause a mental disorder’ and ‘excessive drinking of alcohol can cause a mental disorder’). For the purpose of the present study, mental disorder was translated into Chinese as 心理病 (literally ‘psychological illness’). It was decided to use the term for psychological illness (xing-li-bin – 心理病) rather than mental illness (jing-shen-bin – 粳神病) as the Chinese term for the latter is often considered synonymous with psychosis. The Chinese term for psychological illness is most closely equivalent to the western term ‘mental disorder’. It was reasoned that, by using the term psychological illness, participants would consider psychological disorders in general rather than focusing only on psychosis (for which there is still substantial variation and confusion as to the meaning of the term (e.g. see Furnham & Chan, 2004).

The questionnaire in the present study included 32 statements covering a wide range of possible causes of mental disorders (Table 3). As well as more ‘conventional’ causes (e.g. biological, stress, drugs, loneliness, gambling, relationship problems, bereavement, financial difficulties, etc.) included in western-oriented questionnaires (e.g. Eisenbruch, 1990), the questionnaire also included a range of causes related to ‘traditional’ Chinese culture. These included fate (e.g. ‘people who suffer from a mental disorder are a victim of fate’), ghosts (e.g. ‘ghosts can cause a mental disorder’), karma (‘a mental disorder can be punishment for bad deeds’) and feng shui (‘poor feng shui can cause a mental disorder’). Other possible causes, including excessive thinking (‘thinking too much can cause a mental disorder’) and relationship problems (‘within the family’ and also ‘outside of the family’), which has been identified as a commonly cited cause of mental illness in other studies of Chinese populations (Phillips et al., 2000), were also included in the questionnaire. In addition to exploring the extent of endorsement of different etiology beliefs, the questionnaire also included a measure of general attitudes towards individuals with a mental disorder (a measure of ‘social distance’). It was hypothesized that individuals who more highly endorsed ‘traditional’ beliefs would also indicate more negative attitudes to individuals with a mental disorder, since traditional beliefs are generally considered to contribute to the stigma of mental illness in Chinese culture (Lam et al., 2010).

Method

Participants

A total of 389 questionnaires were returned, but, due to missing data, only those from 361 participants were analysed. Of these, 217 (60.1%) were female and 144 (39.9%) were male (Table 1).

Table 1.

Summary of responses to demographic questions showing the response categories and the number and percentage of responses for each category.

Demographic	Categories	Frequency^a	%
Gender	Male	144	39.9
	Female	217	60.1
First language	Cantonese	339	93.9
	Mandarin	7	1.9
	Other	4	1.1
Marital status	Single	160	44.4
	Married	186	51.7
	Separated	2	0.6
	Divorced	10	2.8
	Widowed	2	0.6
Age	18–25	81	22.8
	26–35	120	33.7
	36–45	50	14.0
	46–55	63	17.7
	56–65	31	8.7
	66+	11	3.1
Highest education level	Primary	32	8.9
	Secondary	145	40.3
	Degree	159	44.2
	Postgraduate degree	24	6.7
Religious beliefs	No	241	66.9
	Yes	119	33.1
	Catholic	29	24.4
	(Other) Christian	34	28.6
	Buddhist	51	42.8
	Taoist	1	0.8
	Other	4	3.4

11 respondents (3%) did not indicate their first language; 1 (0.3%) did not indicate marital status; 5 (1.4%) did not indicate age; 1 (0.3%) did not indicate highest education level; and 1 (0.3%) did not indicate religious beliefs.

Materials

Each participant was given a questionnaire to complete. All information, instructions, questions and statements were written using traditional Chinese characters. The questionnaire was initially written in English and then translated into Chinese using a back-translation procedure to ensure consistency of meaning. The initial translation into Chinese was performed by a native Chinese speaker fluent in English who was a graduate student in psychology and who worked in a clinical setting. The back-translation was performed by a different psychology graduate, also a native Chinese speaker fluent in English, and who also worked in a clinical setting. Minor discrepancies between the original English and back-translated versions were resolved in discussion with the two translators.

Procedures

Samples in previous studies (Luk & Bond, 1992; Mathews, 2011) included only student participants. In an effort to include older participants, two approaches to sampling were employed. A third of the questionnaires (33.5%) were distributed by a researcher in public locations around Macau (mostly public parks and gardens), and were completed by the participants and handed back to the researcher. The remaining questionnaires were distributed within a range of workplaces, organizations and associations in Macau. The participants were given the questionnaires and were requested to return them within, at most, two days. Despite these efforts, the sample obtained was still not entirely representative of the Macau population, with a larger number of female respondents and a high proportion of participants with a higher education (Table 1).

Measures

The questionnaire completed by participants comprised a number of questions and statements organized in several sections, as follows:

Demographics. After some background information explaining the study and its consensual and confidential nature, a short series of demographic questions were presented. These included gender, marital status, age category, highest education level and religious beliefs (Table 1).

Statements related to the etiology of mental disorders. The questionnaire included 32 statements concerning the possible etiology of mental disorders. (For a complete list of all etiology statements see Table 3). For each of the statements, participants were requested to indicate their level of agreement with the statement on a six-point Likert-type scale (1 = ‘strongly disagree’, 2= ‘disagree’, 3 = ‘slightly disagree’, 4 = ‘slightly agree’, 5 = ‘agree’, 6 = ‘strongly agree’).

(Negative) attitudes towards people with a mental disorder. Twelve statements related to general attitudes towards those with a mental disorder (referred to as ‘social distance’, Chung & Chan, 2004; Schulze, Richter-Werling, Matschinger & Angermeyer, 2003). As for the etiology statements, participants were requested to indicate their level of agreement with each statement on a six-point Likert-type scale (1 = ‘strongly disagree’, 2 = ‘disagree’, 3 = ‘slightly disagree’, 4 = ‘slightly agree’, 5 = ‘agree’, 6 = ‘strongly agree’). The twelve statements were: ‘I would be afraid to talk to someone who has a mental disorder’; ‘people with a mental disorder are more likely to attack other people’; ‘I would be uncomfortable to be in the same room with someone who has a mental disorder’; ‘I could be friends with someone who has a mental disorder (*)’; ‘I would feel ashamed if my friends knew that someone in my family had a mental disorder’; ‘people with a mental disorder are dangerous’; ‘if one of my friends developed a mental disorder, I would go and see them (*)’; ‘it is hard to know what someone with a mental disorder will do next’;‘I would not introduce someone who has a mentaldisorder to my friends’; ‘someone who has a mental disorder should not work with children’; ‘I do not want to work with someone with a mental disorder’; and ‘it is not possible to predict how a person with a mental disorder will behave’. Responses to the two statements with asterisks were reverse-scored. When combined, higher-scored responses to these 12 statements would indicate more negative attitudes towards the mentally ill (greater ‘social distance’).

Contact with individuals with a mental disorder. Five statements related to participants’ previous contact with individuals with a mental disorder (Table 2). Participants responded either ‘true’ or ‘false’ to each statement. These statements are based on the Level of Contact report items used by Corrigan, Edwards, Green, Diwan and Penn (2002), but considerably simplified. A greater number of ‘true’ responses indicates a greater degree of contact with individuals with a mental disorder.

Other attitudes related to mental illness. Three statements related to participants’ general attitudes to mental disorders (‘it is possible to treat a person with mental disorder’; ‘a mental disorder is like any other illness’; ‘I would like to have more knowledge about mental disorders’). Participants were requested to indicate their level of agreement with these statements on the same six-point Likert-type scale used for the etiology statements.

Table 2.

Summary of responses to items in the questionnaire related to the level of contact that participants have had with someone with a mental disorder.

Questionnaire item related to contact of participant with someone with a mental disorder	Response
	True n (%)	False n (%)
I have seen someone with a mental disorder on TV or in a film	330 (92.2)	28 (7.8)
I have seen someone with a mental disorder in real life	281 (78.9)	75 (21.1)
I know someone with a mental disorder	161 (45.2)	195 (54.8)
I have spoken to someone with a mental disorder	197 (55.5)	158 (44.5)
I have a relative with a mental disorder	75 (21.1)	281 (78.9)

Statistical analysis

A total of 389 questionnaires were returned but several included missing data (where respondents had not indicated a response to a questionnaire item). An analysis of this missing data revealed that 281 questionnaires had no missing data. Of the remaining 108 questionnaires, the majority had only one or two missing values. To exclude questionnaires with excessive amounts of missing data, 28 questionnaires with more than 5% missing values (more than three missing values) were excluded from the analysis. Therefore, data from a total of 361 respondents were analysed. However, for some of the analyses presented, the numbers of respondents may be less due to the presence of some missing data for some respondents.

All data were entered in SPSS 19.0 for analysis, which included the calculation of descriptive statistics, correlation, reliability and factor analysis as well as comparing scores with inferential tests (analysis of variance (ANOVA)).

Results

Table 1 shows a summary of responses to demographic items for the participants in the present study. The results indicate a larger number of female participants (60.1%) compared with males (39.9%). The vast majority of participants indicated that they were either single (44.4%) or married (51.7%). In terms of age, only a small number of participants were aged over 55, with the largest age group being 26–35 years of age. Most of the participants had completed at least secondary education, with many indicating that they had completed higher education. Approximately a third of participants indicated having religious beliefs, predominately Buddhist and Christian (including Catholic) beliefs. As would be expected, the vast majority of respondents indicated that their first language was Cantonese.

Table 2 shows the responses to items related to the participants’ contact with individuals who have a mental disorder. The vast majority of participants (92.2%) indicated that they had ‘seen someone with a mental disorder on TV or in a film’. A smaller percentage (78.9%) had seen someone with a mental disorder in real life, and fewer still had spoken with someone with a mental disorder (55.5%). Perhaps unexpectedly, quite a large percentage of participants (45.2%) indicated that they knew someone with a mental disorder, and more than 20% of participants indicated that they had a relative with a mental disorder.

The 32 questionnaire items that related to the etiology of mental disorders required participants to indicate how strongly they agreed with each statement and mean scores are shown in Table 3. For all participants, the statement that ‘stress can cause mental disorder’ was most strongly endorsed. This was also the case when female and male participants were considered separately. The statement that all participants agreed with the least was ‘if someone develops a mental disorder it is mostly chance’. Again, this was also true for female and male participants separately.

Table 3.

Mean response scores for each of the 32 etiology-related statements in the questionnaire.

Rank	Etiology statement	Mean agreement
1	Stress can cause a mental disorder	4.99
2	Too much stress increases the risk of developing a mental disorder	4.83
3	Drug use can cause a mental disorder	4.68
4	Anyone can get a mental disorder	4.60
5	Problems in family relationships can cause a mental disorder	4.43
6	Money worries can cause a mental disorder	4.36
7	Thinking too much can cause a mental disorder	4.35
8	Excessive gambling can cause a mental disorder	4.24
9	A death in the family can cause a mental disorder	4.12
10	Health problems can cause a mental disorder	4.06
11	Problems in relationships with people outside the family can cause a mental disorder	4.04
12	Excessive drinking of alcohol can cause a mental disorder	4.02
13	People with financial difficulties are more likely to develop a mental disorder	4.01
14	A mental disorder can be the result of inheritance	3.88
15	Personality problems can lead to a mental disorder	3.75
16	Loneliness can cause a mental disorder	3.73
17	People with good physical health are less likely to develop a mental disorder	3.63
18	Guilt can lead to mental disorder	3.49
19	People who spend a lot of time alone are more at risk of developing a mental disorder	3.34
20	Feelings of shame can cause a mental disorder	3.13
21	A sickness or infection can cause a mental disorder	3.09
22	Lack of education increases the risk of a mental disorder	3.07
23	People who suffer from a mental disorder are a victim of fate	2.98
24	Physically weak people are more at risk of developing a mental disorder	2.97
25	A poor upbringing can cause a mental disorder	2.82
26	Ghosts can cause a mental disorder	2.74
27	A poor education can lead to mental disorder	2.66
28	If you spend time with people who have a mental disorder there is a risk of suffering a mental disorder	2.55
29	A mental disorder can be punishment for bad deeds	2.46
30	It is the fate of some people to suffer from a mental disorder	2.37
31	Poor feng shui can cause a mental disorder	2.26
32	If someone develops a mental disorder it is mostly chance	2.14

Note: A higher score indicates greater agreement with the statement. Mean scores are shown for all participants, with statements ranked from highest to lowest agreement. A mean score > 4 indicates general agreement with the statement (4 = ‘slightly agree’, 5 = ‘agree’, 6 = ‘strongly agree’). A mean score between 3 and 4 approximates to a neutral attitude towards the statement, and a score < 3 indicates general disagreement with the statement (1 = ‘strongly disagree’, 2 = ‘disagree’, 3 = ‘slightly disagree’).

For all participants, the second-ranked statement related to stress, with the third most endorsed statement relating to drug use as a possible cause of mental disorder. These two statements also made the top three ranked statements for females and males separately, although males indicated slightly greater agreement with drugs as a cause (ranked second instead of third). For the fourth-ranked statement, participants indicated agreement with the statement that ‘anyone can get a mental disorder’. The remaining statements that participants agreed with (mean score greater than 4) as likely causes of mental disorders included: relationship problems, especially within the family; financial problems; gambling and alcohol problems; a death in the family; and thinking too much.

When female and male participants were considered separately, some differences were evident for the statements that participants generally agreed with. For example, males tended to agree more with both money worries and financial difficulties as possible causes of mental disorder compared with females (4.49 vs 4.27 and 4.20 vs 3.88, respectively, for males vs females). Males also tended to agree more that health problems (4.15 compared with 4.00 for females) and excessive drinking of alcohol may lead to a mental disorder (4.13 compared with 3.94 for females). For female participants, there was greater agreement that ‘thinking too much’ is a cause of mental disorder (4.45 compared with 4.19 for males).

Statements concerning the cause of mental disorders that participants were, overall, relatively neutral towards (mean scores between 3 and 4 for all participants) included such causes as physical health (including infections), inheritance, a lack of education and psychological factors (personality, loneliness, guilt and shame). No noticeable gender differences in how much these statements were endorsed was evident.

Statements concerning the cause of mental disorders that participants, overall, disagreed with (mean scores below 3) included such causes as physical weakness, a poor education/upbringing, spending time with the mentally ill and traditional Chinese beliefs. These results indicate that participants generally disagreed with fate, ghosts, bad deeds and poor feng shui as causes of mental disorder. Such traditional beliefs relating to the causes of mental disorders are generally ranked below all other possible causes, being ranked 23rd/30th, 26th, 29th and 31st, respectively out of the 32 statements presented.

To explore whether such traditional beliefs cluster together as a meaningful component, or underlying factor, in peoples’ beliefs about the cause of mental disorders, responses to the 32 etiology statements were entered into an exploratory factor analysis. To screen the data for factor analysis, a correlation matrix between responses to all 32 items was calculated. For factor analysis, it is important that there is some, but not too much, correlation between the variables (responses to the statements). The correlation matrix indicated that no variables were highly correlated. However, responses to some statements did not correlate moderately (correlation coefficient > 0.3) with responses to any other statements (responses to all other statements correlated moderately with responses to at least three other statements). These statements were those ranked 4th, 14th, 15th and 17th as possible causes of mental disorder(anyone can get a mental disorder, physical health, inheritance and personality problems). Responses to these four statements were excluded from the factor analysis.

The remaining 28 statements were factor-analysed (principal components analysis (PCA) with varimax rotations). The overall Kaiser-Meyer-Olkin (KMO) measure of sample adequacy was 0.844, in the range generally considered ‘good’ for factor analysis. This was confirmed by Bartlett’s test of sphericity, which was highly significant(p < .001). As the mean communalities score was 0.601 and the sample size was 361, Kaiser’s criteria for determining factors was used (eigenvalues greater than 1). This extracted a seven-factor solution that converged after 17 iterations of rotation and accounted for 60.15% of variance in responses. The factor loadings of the 28 questionnaire items on these seven factors are shown in Table 4.

Table 4.

Rotated factor analysis solution for responses to 28 etiology statements showing the factor loading of each statement on the seven factors extracted.

Etiology statement	Extracted components/factors
	1	2	3	4	5	6	7
People with financial difficulties are more likely to develop a mental disorder	.460
A poor upbringing can cause a mental disorder					.789
Drug use can cause a mental disorder				.783
Excessive drinking of alcohol can cause a mental disorder				.749
Problems in relationships with people outside the family can cause a mental disorder							–.492
Stress can cause a mental disorder	.503			.453
Lack of education increases the risk of a mental disorder					.721
A mental disorder can be punishment for bad deeds					.413	.512
Thinking too much can cause a mental disorder	.674
A death in the family can cause a mental disorder	.643
Problems in family relationships can cause a mental disorder	.756
Health problems can cause a mental disorder	.442
Money worries can cause a mental disorder	.637
People who suffer from a mental disorder are a victim of fate						.696
Ghosts can cause a mental disorder			.576
If you spend time with people who have a mental disorder there is a risk of suffering a mental disorder			.781
Poor feng shui can cause a mental disorder			.752
Feelings of shame can cause a mental disorder		.609
If someone develops a mental disorder it is mostly chance			.665			.435
Physically weak people are more at risk of developing a mental disorder		.425					.542
People who spend a lot of time alone are more at risk of developing a mental disorder		.467					.508
Excessive gambling can cause a mental disorder		.405		.655
Too much stress increases the risk of developing a mental disorder				.452
A poor education can lead to mental disorder					.639
Guilt can lead to mental disorder		.689
A sickness or infection can cause a mental disorder		.617
Loneliness can cause a mental disorder	.429	.658
It is the fate of some people to suffer from a mental disorder						.633

Note: Only factor loadings greater than .4 or less than -.4 are shown. The seven statements related to ‘traditional (Chinese) beliefs’ are shown in italics. Scores for responses to these seven statements were combined to create a measure referred to as a traditional beliefs score.

Table 4 shows that statements that relate to more traditional Chinese beliefs clustered on two of the seven factors. Ghosts, spending time with the mentally ill, poor feng shui and chance all loaded most strongly on factor 3 (‘chance’ also loaded on factor 6). The statements for causes of mental disorder due to punishment for bad deeds, being a victim of fate and it being the fate of some people all loaded most strongly on factor 6. These results indicate that statements concerning traditional causes of mental disorder relate to two underlying factors (factors 3 and 6 in Table 4). Factor 3 seems to relate to spirits/forces and factor 6 to fate/chance. However, responses to the statement ‘if someone develops a mental disorder it is mostly chance’ did load most strongly on factor 3 (while also loading on factor 6), so this separation does not seem to be completely ‘clean’.

Based on the results of the factor analysis, it is reasonable to combine responses to the seven statements related to ‘traditional beliefs’ into a single measure. Responses to these seven items were summed to calculate a new measure referred to as ‘traditional beliefs score’. A reliability analysis of this traditional beliefs score indicated moderate reliability for a measure comprised of only seven items (Cronbach’s α = 0.79). Removing any one of these seven items from this measure resulted in a reduction in internal consistency.

Table 5 shows mean traditional beliefs scores for several demographic groups of participants. For nearly all the demographics, no difference in mean traditional beliefs scores was evident, except for education. Participants without higher education (most of whom had completed secondary school) had a noticeably higher mean score compared with participants who had completed higher education (18.43 compared with 16.43). A series of ANOVA confirmed this observation, with the only significant difference related to the education level of participants. Participants without a higher education had significantly higher mean traditional beliefs scores compared with those who had completed a higher education degree (F(1, 344) = 10.130, p = .002).

Table 5.

Mean traditional beliefs scores (the extent to which participants agreed with statements related to ‘traditional causes’ of mental disorders) related to different demographic groups.

Demographic	Categories	Mean traditional beliefs score	Significance of difference (p) by one-way ANOVA
Gender	Male (n = 134)Female (n = 213)	17.4817.39	.893
Marital status	Single (n = 155)Married (n = 178)	17.4517.20	.700
Age	Up to 35 years (n = 197)36 years and above (n = 145)	17.2017.72	.418
Highest education level	No higher education (n = 167)Higher education (n = 179)	18.4316.43	.002**
Religious beliefs	No (n = 230)Yes (n = 116)	17.4017.39	.991

p < .01.

Twelve of the remaining statements in the questionnaire were intended to measure the participants’ attitudes towards the mentally ill (the ‘social distance’ between participants and individuals with a mental disorder). Scores for these statements were combined to give an overall score of ‘social distance’ (scores for two statements were reverse-scored). However, a reliability analysis of responses to these 12 statements revealed that inclusion of responses to the statement ‘if one of my friends developed a mental disorder, I would go and see them’ actually lowered the internal consistency (Cronbach’s α) of the overall measure. Presumably the directly personal reference in this statement meant that responses to it were not consistent with responses to the other 11 statements. Responses to this statement were therefore excluded from the measure of social distance score. The internal consistency of responses to the remaining 11 statements combined to create the measure social distance was acceptable (Cronbach’s α = 0.81).

Table 6 shows mean social distance scores for different demographic groups of participants. In terms of gender, both males and females were very similar in terms of social distance scores. Similarly, no difference in scores was evident related to whether participants had religious beliefs or not. However, for three demographics (marital status, age and education), a significant difference was evident. Of these, the largest difference was between younger and older participants, with significantly higher social distance scores for older participants (F(1,336) = 18.721, p < .001). With respect to marital status, social distance scores for married participants were significantly higher than for single participants (F(1,329) = 9.286, p = .002). For education, participants with a higher education degree had lower social distance scores compared with participants who had not completed higher education (F(1,341) = 9.926, p = .002).

Table 6.

Mean social distance scores (higher scores indicating a greater desire for social distance and so a more negative attitude toward the mentally ill) related to different demographic groups.

Demographic	Categories	Mean social distance score	Significance of difference (one-way ANOVA) p
Gender	Male (n = 134)Female (n = 208)	38.2538.03	.798
Marital status	Single (n = 155)Married (n = 174)	36.6939.26	.002**
Age	Up to 35 years (n = 193)36 years and above (n = 145)	36.5540.13	< .001**
Highest education level	No higher education (n = 167)Higher education (n = 179)	39.5536.94	.002**
Religious beliefs	No (n = 232)Yes (n = 109)	37.9238.57	.469

p < .01.

The relationship between the three measures derived from the questionnaire responses (level of contact scores, traditional beliefs scores and social distance scores) were investigated using correlation. Table 7 shows the correlation coefficients (Spearman’s rho) for pairs of these measures. No correlation was evident between the level of contact scores and the traditional beliefs scores (r = .023). However, the level of contact scores was correlated mildly with social distance scores (r = -0.179). This significant negative correlation indicates that as level of contact scores increase, social distance scores decrease. So the more contact participants reported with people with mental disorders, the less negative attitudes they expressed towards the mentally ill. The strongest correlation was between the traditional beliefs scores and the social distance scores of participants (r = 0.375). This significant correlation indicates a moderately strong positive relationship between endorsing traditional beliefs about the etiology of mental disorders and indicated social distance. Therefore, as participants agreed more with traditional causes of mental disorders (e.g. ghosts, fate, etc.), they also tended to hold more negative attitudes towards the mentally ill.

Table 7.

Correlation coefficients (Spearman’s rho) between the three measures derived from questionnaire responses.

	Level of contact scores	Traditional beliefs scores	Social distance scores
Level of contact scores	1.000	0.023p = .685	−0.179**p = .001
Traditional beliefs scores	–	1.000	0.375**p < .001
Social distance scores	–	–	1.000

p < .01.

Finally, the questionnaire included three statements that asked participants about their general attitudes to mental disorder statements (‘it is possible to treat a person with mental disorder’; ‘a mental disorder is like any other disorder’; ‘I would like to have more knowledge about mental disorder’). Responses to these statements indicated that participants slightly agreed that a mental disorder is like any other disorder (mean response = 4.00). Participants indicated that they agreed that it is possible to treat a mental disorder (mean response = 4.90) and also that they would like to have more knowledge about mental disorders (mean response = 4.47).

Discussion

The results of the present study indicate that traditional Chinese beliefs concerning the cause of mental disorders were perceived as relatively less important than psychosocial causes. The endorsement of traditional beliefs was significantly correlated with an increased desire for social distance from individuals with a mental disorder. The results also indicated significantly lower levels of endorsement of traditional beliefs for participants with higher levels of education.

As in the study of Phillips et al. (2000), stress was perceived as the most important cause of a mental disorder, followed by other psychosocial causes such as drug use, money worries and relationship problems. Also consistent with the findings of Phillips et al., excessive thinking was perceived as a relatively important cause of mental disorders (ranked 7th out of 32). However, unlike the results of Phillips et al. which ranked personality problems as the most important cause of mental illness, in the present study personality problems were not ranked as especially important (ranked only 15th out of 32). This divergence may be related to either methodological differences between the two studies or differences in the participants. In Phillips et al., the mainland China participants were asked to indicate the cause of the illness suffered by a family member. The present study surveyed public attitudes in Macau and participants were asked to indicate agreement with all possible causes rather than the more complex process through which participants endorsed causes in the study of Phillips et al. (2000).

In the present study, the term mental disorder (xing-li-bin – 心理病) rather than mental illness (jing-shen-bin – 精神病) was used to assess attitudes related to etiology. Terminology is an important factor in studies of attitudes and beliefs concerning the causes and nature of mental illness/disorders (Found & Duarte, 2011). However, it is difficult to interpret the meaning of different terms for the general public. The results provide a clue as to how the participants perceived the meaning of the term mental disorder (xing-li-bin – psychological illness) used in the present study. When asked about their contact with individuals with a mental disorder, 21% of participants indicated that they had a relative with a mental disorder. Such a high figure indicates that participants did indeed interpret the meaning of the term mental disorder (xing-li-bin – psychological illness) in the more general sense intended. It is unclear whether participants would perceive different causes for psychosis (jing-shen-bin – mental illness) compared with mental disorders more generally (xing-li-bin – psychological illness).

Although the present results indicated a relatively low importance for traditional Chinese beliefs about the causes of mental illness, the factor analysis supported the idea of such beliefs as a distinct class or group of causes. This finding is consistent with the results of Mathews (2011). Mathews identified two factors, referred to as ‘supernaturally based’: one related to fate and karma (e.g. ‘astrological destiny’, ‘bad luck’), and the other to religious concepts and supernatural forces (e.g. ‘demonic forces’, ‘sin’). This distinction is supported in the results of the present study. The factor analysis of responses to the etiology statements of the present study also identified two factors that ‘traditional causes’ loaded heavily on (factors 3 and 6). Factors 3 and 6 are similar to the ‘religious’ and ‘fate’ factors of Mathews’ study.

In both the results of the present study and the study of Mathews (2011), the question of whether certain etiology causes are really ‘supernatural’ is an important issue. It is a matter of interpretation as to whether ‘traditional’ beliefs can be seen as synonymous with supernatural or superstitious beliefs (and also whether some of the causes studied here are uniquely Chinese). For example, the statement in the present study ‘a mental disorder can be punishment for bad deeds’ is related to the traditional belief of karma, derived from Buddhism, which is an important influence in Chinese culture. However, such a belief is not supernatural and need not necessarily be interpreted as superstitious. Instead, it is psychologically plausible that regret and shame concerning past behaviour could manifest as a mental disorder. It may therefore be more appropriate to refer to such beliefs as simply ‘cultural’ rather than using value-laden terms such as ‘supernatural’ (or even ‘traditional’). The very fact that some individuals believe in certain causes of mental disorders means that they have relevance to their disorder and this may indeed contribute to the development of a mental disorder for these individuals. In the context of mental disorders, what people believe matters, whether it is scientifically valid or not (Bhui & Bhugra, 2002; Halligan, 2007).

The participants in the present study were the general public of Macau. Although they were not asked about their current psychological health, it is reasonable to assume that only a minority of participants were suffering from a psychological disorder at the time. Etiology beliefs have also been studied in clinical populations. For example, McCabe and Priebe (2004) compared etiology beliefs among four ethnic groups of adult schizophrenia outpatients in the UK. Participants completed a modified version of the Short Explanatory Model Interview (SEMI; Lloyd et al., 1998) as well as other psychological measures. Although only a minority of participants considered their disorder to have a supernatural cause, this was higher than those who indicated a biological cause. This indicates that perceptions of supernatural causes of mental disorder may be more common in clinical populations. When responses from participants from the four ethnic groups were compared, there were significant differences in the frequency with which supernatural causes were cited, with no white participants citing supernatural causes compared to 18 participants from other ethnic groups who did cite supernatural causes for their disorder. This result is consistent with cultural background as an important influence on both the patient’s and the general public’s explanatory models of illness (Prior, Chun & Huat, 2000; Yang et al., 2009). Further research into cultural differences in etiology beliefs in both non-clinical and clinical populations, as well as the psychological consequences (e.g. social distance) and utility of such beliefs (e.g. stigma reduction of sufferers), is needed.

Footnotes

Acknowledgements

Thanks to Sin U Leong, Chi Hou Ip, Iok Peng Chim and Nga Si Ho for their hard work collecting data for the present study. Also, thanks to Sin U Leong and Yu Wai Chong for their patient assistance in the translation of material between English and Chinese.

References

Abe-Kim

Takeuchi

Hong

Zane

Sue

Spencer

M. S.

Alegría

(2007). Use of mental health-related services among immigrant and US-born Asian Americans: Results from the National Latino and Asian American Study. American Journal of Public Health, 97, 91–98.

Angermeyer

M. C.

Matschinger

(1996). Public attitude towards psychiatric treatment. Acta Psychiatrica Scandinavica, 94, 326–336.

Atkinson

D. R.

Worthington

R. L.

Dana

D. M.

Good

G. E.

(1991). Etiology beliefs, preferences for counseling orientations, and counseling effectiveness. Journal of Counseling Psychology, 38, 258–264.

Bhui

Bhugra

(2002). Explanatory models for mental distress: Implications for clinical practice and research. British Journal of Psychiatry, 181, 6–7.

Callan

Littlewood

(1998). Patient satisfaction: Ethnic origin or explanatory model? International Journal of Social Psychiatry, 44, 1–11.

Chen

A. W.

Kazanjian

Wong

(2009). Why do Chinese Canadians not consult mental health services: Health status, language or culture? Transcultural Psychiatry, 46, 623–641.

Chen

X. S.

Mak

W. S.

(2008). Seeking professional help: Etiology beliefs about mental illness across cultures. Journal of Counseling Psychology, 55, 442–450.

Chu

J. P.

Sue

(2011). Asian American mental health: What we know and what we don’t know. Online Readings in Psychology and Culture, Unit 3. Retrieved from http://scholarworks.gvsu.edu/orpc/vol3/iss1/4

Chung

K. F.

Chan

J. H.

(2004). Can a less pejorative Chinese translation for schizophrenia reduce stigma? A study of adolescents’ attitudes toward people with schizophrenia. Psychiatry and Clinical Neuroscience, 58, 507–515.

10.

Corrigan

P. W.

Edwards

Green

Diwan

S. L.

Penn

D. L.

(2002). Prejudice, social distance, and familiarity with mental illness. Schizophrenia Bulletin, 27, 219–225.

11.

Eisenbruch

(1990). Classification of natural and supernatural causes of mental distress: Development of a Mental Distress Explanatory Model. Journal of Nervous and Mental Disease, 178, 712–719.

12.

Found

Duarte

(2011). Attitudes to mental illness: The effects of labels and symptoms. East Asian Archives of Psychiatry, 21, 91–98.

13.

Furnham

Bower

(1992). A comparison of academic and lay theories of schizophrenia. British Journal of Psychiatry, 161, 201–210.

14.

Furnham

Chan

(2004). Lay theories of schizophrenia: A cross cultural comparison of British and Hong Kong Chinese attitudes, attributions and beliefs. Social Psychiatry and Psychiatric Epidemiology, 39, 543–552.

15.

Furnham

Henley

(1988). Lay beliefs about overcoming psychological problems. Journal of Social and Clinical Psychology, 26, 423–438.

16.

Gunn

G. C.

(1996). Encountering Macau: A Portuguese city-state on the periphery of China, 1557-1999. Boulder, CO: Westview.

17.

Halligan

(2007). Belief and illness. The Psychologist, 20, 358–361.

18.

Holzinger

Loffler

Muller

Priebe

Angermeyer

M. C.

(2002). Subjective illness theory and compliance of schizophrenia patients with neuroleptic treatment. Journal of Nervous and Mental Disease, 190, 597–603.

19.

Kelly

G. R.

Mamon

J. A.

Scott

J. E.

(1990). Utility of the health belief model in examining medication compliance among psychiatric outpatients. Social Science and Medicine, 25, 1205–1211.

20.

Kleinman

(1980). Patients and healers in the context of culture: An exploration of the borderland between anthropology, medicine and psychiatry. Berkeley, CA: University of California Press.

21.

Lam

C. S.

Tsang

H. W. H.

Corrigan

P. W.

Lee

Y. T.

Angell

Shi

Larson

J. E.

(2010). Chinese lay theory and mental illness stigma: Implications for research and practices. Journal of Rehabilitation, 76, 35–40.

22.

Le Meyer

Zane

Cho

Y. I.

Takeuchi

D. T.

(2009). Use of specialty mental health services by Asian Americans with psychiatric disorders. Journal of Consulting and Clinical Psychology, 77, 1000–1005.

23.

Lloyd

K. R.

Jacob

K. S.

Patel

St. Louis

Bhugra

Mann

A.H.

(1998). The development of the Short Explanatory Model Interview (SEMI) and its use among primary-care attenders with common mental disorders. Psychological Medicine, 28, 1231–1237.

24.

Luk

C.-L.

Bond

M. H.

(1992). Chinese lay beliefs about the causes and cures of psychological problems. Journal of Social and Clinical Psychology, 11, 140–157.

25.

McCabe

Priebe

(2004). Explanatory models of illness in schizophrenia: Comparison of four ethnic groups. British Journal of Psychiatry, 185, 25–30.

26.

Mathews

(2011). Assessment and comparison of culturally based explanations for mental disorder among Singaporean Chinese youth. International Journal of Social Psychiatry, 57, 3–17.

27.

MSCS (Macau Statistics and Census Service). (2007). Global results of by-census 2006. Available at: http://www.dsec.gov.mo

28.

Norton

J. P.

(2007). Depression Anxiety and Stress Scales (DASS-21): Psychometric analysis across four racial groups. Anxiety, Stress & Coping, 20, 253–265.

29.

Phillips

M. R.

Stroup

S. T.

Xin

(2000). Causes of schizophrenia reported by patients’ family members in China. British Journal of Psychiatry, 177, 20–25.

30.

Prior

Chun

P. L.

Huat

S. B.

(2000). Beliefs and accounts of illness: Views from two Cantonese-speaking communities in England. Sociology of Health & Illness, 22, 815–839.

31.

Saravanan

Jacob

K. S.

Johnson

Prince

Bhugra

David

A. S.

(2007). Belief models in first episode schizophrenia in South India. Social Psychiatry and Psychiatric Epidemiology, 42, 446–451.

32.

Schulze

Richter-Werling

Matschinger

Angermeyer

M. C.

(2003). Crazy? So what! Effects of a school project on students’ attitudes towards people with schizophrenia. Acta Psychiatrica Scandinavica, 107, 142–150.

33.

Yang

L. H.

Phillips

M. R.

Chou

Zhang

Hopper

(2009). ‘Excessive thinking’ as explanatory model for schizophrenia: Impacts on stigma and ‘moral’ status in mainland China. Schizophrenia Bulletin, 36, 836–845.

34.

Young

C. B.

Fang

D. Z.

Zisook

(2010). Depression in Asian-American and Caucasian undergraduate students. Journal of Affective Disorders, 125, 379–382.