Quantitative Analysis of Healthy Ageing in England - Assignment Example

Quantitative analysis of healthy ageing in England By Department 04 April Introduction Today life expectancy in has been increasing in UK in recent history covering past 50 years and by 2050 the expectancy age will be 91 years. In UK as today we have over 15,000 population of senior citizen aged 100 years and above (Crown copyright, 2014). The retirement age has become more active phase in life where citizens are responsible for looking at their pension and health. We commission a study to evaluate the dynamics of the senior citizens of UK aged between 50 to 99 years, in regard to their health and income. Literature review World Health Organization has been promoting a healthy ageing as people’s ability regardless of age to have a healthy living, socially inclusive lifestyle and safe living. Some factors surpass social care and health that have a major effect on person’s well-being and health, this embraces a health reproach to life course in that it has an impact on life experiences on a population age group set (Age UK, n.d.). Ageing increases frailty and as a result prevention of illness and disability management are key toward promoting a culture of healthy ageing. Thus it’s prudent by design to prevent or delay various chronic diseases thus increasing life expectancy; this is influenced by individual locus of responsibilities on socio-economic factors, cultural influences factors and environmental impact on the age group (Age UK, n.d.). Physical exercises play a vital role in reduction of and prevention of chronic diseases and combating age-related illness. Senior citizens faces a barrage of issues not confined in psychological and physiological, but in addition depression, loneliness and isolation even loss of mobility and independence (Mary Kate Connolly, n.d.). Physical mobility has a great impact in elderly vulnerable population group, it increases the quality of life and ability to socialize thus killing loneliness. Depression prevalence in aged people can be linked to lack to regular physical activity. Physical activity involvement in older people promotes positive perceptions in psychological well-being. Thus, psychological well-being becomes the main predictor for remaining physically active in old age as, as relationship between mental and physical health are intertwined and interrelated (Linda Seymour, 2004). Research question Is depression and chronic ailments are caused due to negligence of senior citizens in England? Hypothesis Ho = Depression is caused by prevalence of disability. Ha = Depression is not caused by prevalence of disability. Ho = Income has an effect on healthy aging in England. Ha = Income has no effect on healthu aging in England. Data analysis The data is secondary data obtained of ELSA. We shall use statistical package SPSS to analyse our data. 1. Health conditions Main variables include the following:- Sex of the respondent’s a qualitative variable with nominal attribute. Age it’s a quantitative variable with continuous attribute. Income is a quantitative variable with discrete variable. Chronic Ailments is a qualitative variable with nominal attributes. Limitations Disability is a quantitative variable with discrete attribute. The Chronic Ailments is a variable generated from collapsing multiple variables to form one variable. This chronic ailments include the heart conditions and the chronic lung diseases. The non-communicable diseases allied to heart condition include hypertension, angina, diabetes, stroke and heart ailments including heart attack, congestive heart failure, heart murmur, abnormal heart rhythm and other conditions. The chronic lung diseases include emphysema, asthma, arthritis, osteoporosis, cancer, Parkinson, dementia, Alzheimer’s, and emotional problems. The new variable chronic ailments derives the total number of the health conditions to give the new variable a parametric nature that of a continuous variable and has a population of a normal distribution (Belinda Barton, 2014). Thus by recode the variables and summing them up we generate a new variable as shown in the appendix 1 on recode 1.1 The Limitation disability as a new variable of prevalence of disabilities that limits the activities of old age for a healthy age. It is derived by combining multiple variables to form a quantitative continuous variable that can be subjected to a parametric test; thus it as a normal distribution (Triola, 2015). The new variable is generated as shown in the appendix 1 on 1.2 Descriptive statistics Table 1.0 indicates that 56% of the respondents were females and 44% were males as illustrated further by the graph below. Source: ELSA Table 1.0. Frequency Percent Cumulative Percent Valid Male 5335 44.1 44.1 Female 6764 55.9 100.0 Total 12099 100.0 The average senior age is 64 years, the age is a skewed to the right at +.4 an indication that the distribution is approximately symmetric; thus the values are more concentrated to the mean (Coolican, 2014). From the graph above married of 50 years above consists of about 60% as shown in the Table 1.3 and less than 5% of the ELSA respondents are separated legally. Table 1.2 Descriptive Statistics N Range Mean Std. Deviation Variance Skewness Statistic Statistic Statistic Statistic Statistic Statistic Std. Error Age variable from HH grid collapsed at 90 plus 12099 79 64.19 11.103 123.279 .401 .022 BU equivalised total income 11977 11857.82 248.1142 280.47729 78667.513 15.991 .022 The total mean income is £248, and the data is normally distributed with skewness to the right = +16 as shown in Table 1.2. Table 1.3 Current legal marital status Frequency Percent Valid Percent Cumulative Percent Valid Single, that is never married 672 5.6 5.6 5.6 Married, first and only marriage 6812 56.3 56.3 61.9 Remarried, second or later marriage 1349 11.1 11.2 73.0 Legally separated 144 1.2 1.2 74.2 Divorced 1116 9.2 9.2 83.4 Widowed 2003 16.6 16.6 100.0 Total 12096 100.0 100.0 Total 12099 100.0 Table 1.4 Do you have any long-standing illness, disability or infirmity? ...I mean anything that has troubled you or is likely to affect you over a period of time. Frequency Percent Valid Percent Cumulative Percent Valid Yes 6538 56.7 56.8 56.8 No 4975 43.2 43.2 100.0 Total 11513 99.9 100.0 Table 1.4 shows at indicates 57% of senior citizens have reported being sick or disabled as further clarified by the graph below. Source: ELSA Table 1.5 Descriptives ASK OR CODE RESPONDENT~S SEX Statistic Std. Error Age variable from HH grid collapsed at 90 plus Male Mean 64.79 .139 95% Confidence Interval for Mean Lower Bound 64.52 Upper Bound 65.06 5% Trimmed Mean 64.34 Median 64.00 Variance 101.069 Std. Deviation 10.053 Minimum 50 Maximum 99 Range 49 Interquartile Range 16 Skewness .559 .034 Kurtosis -.318 .068 Female Mean 65.43 .136 95% Confidence Interval for Mean Lower Bound 65.16 Upper Bound 65.70 5% Trimmed Mean 64.94 Median 64.00 Variance 116.815 Std. Deviation 10.808 Minimum 50 Maximum 99 Range 49 Interquartile Range 17 Skewness .567 .031 Kurtosis -.337 .062 Table 1.5 shows the females have a mean of 65.4 compared to male 64.7 years of the proportion sample of 50 and above years. Both the sample proportions of males and females are normally distributed with both skewed to the right. Likewise both males and females have a a median of 64 years indicating that most population of in England of older age have an age of 64. The graph shows the age of males are normally distributed and skewed to the right. The sample proportion of females are normally distributed as skewed to the right as shown in the graph. Table 1.6 ELSA ethnic group collapsed into White and Non-white to avoid disclosure Frequency Percent Valid Percent Cumulative Percent Valid White 4858 42.2 97.0 97.0 Non-white 149 1.3 3.0 100.0 Total 5007 43.5 100.0 Missing Refusal 9 .1 Dont know 1 .0 Not applicable 6505 56.5 Total 6515 56.5 Total 11522 100.0 The Table 1.6 shows 56.5% of ELSA respondents did not affiliate with either ethnic group, and 42.2% of the respondents are white. Source : ELSA The graph shows that 97% of the respondents who stated their ethnic group were white. Table 1.7 Descriptive Statistics N Minimum Maximum Mean Std. Deviation BU equivalised total income 11417 -610.00 11247.82 245.3813 281.30456 Valid N (listwise) 11417 Table 1.8a Descriptives ASK OR CODE RESPONDENT~S SEX Statistic Std. Error BU equivalised total income Male Mean 259.7668 3.97151 95% Confidence Interval for Mean Lower Bound 251.9810 Upper Bound 267.5526 5% Trimmed Mean 231.2548 Median 202.4551 Variance 81924.443 Std. Deviation 286.22446 Minimum -497.12 Maximum 11247.82 Range 11744.93 Interquartile Range 182.54 Skewness 15.588 .034 Kurtosis 484.577 .068 Female Mean 233.3746 3.50610 95% Confidence Interval for Mean Lower Bound 226.5014 Upper Bound 240.2477 5% Trimmed Mean 206.3781 Median 177.2742 Variance 76497.623 Std. Deviation 276.58204 Minimum -610.00 Maximum 11247.82 Range 11857.82 Interquartile Range 168.50 Skewness 17.352 .031 Kurtosis 550.902 .062 The average income of senior citizens in England is £ 245 with men having a higher income than women with a mean income of £260 and £233 respectively as shown in the Table 1.7 and 1.8. The income data is normally distributed with data skewed to the right. Table 1.8b Descriptives ASK OR CODE RESPONDENT~S SEX Statistic Std. Error Prevalence Limitation of disability Male Mean 10.6906 .05118 95% Confidence Interval for Mean Lower Bound 10.5903 Upper Bound 10.7910 5% Trimmed Mean 11.2118 Median 12.0000 Variance 13.701 Std. Deviation 3.70146 Minimum .00 Maximum 12.00 Range 12.00 Interquartile Range .00 Skewness -2.525 .034 Kurtosis 4.411 .068 Female Mean 10.7088 .04597 95% Confidence Interval for Mean Lower Bound 10.6187 Upper Bound 10.7990 5% Trimmed Mean 11.2320 Median 12.0000 Variance 13.298 Std. Deviation 3.64657 Minimum .00 Maximum 12.00 Range 12.00 Interquartile Range .00 Skewness -2.561 .031 Kurtosis 4.624 .062 Table 1.8b indicates that mean of females and males are equal at 10.7 with the data distributed normally with skewness = -2.6 indicating the skewness to the left. Table 1.9 Summary of Physical Exercise Count Do you take part in sports or activities that are vigorous ….? {frequency in daily life} ... more than once a week, 2022 once a week, 1085 one to three times a month, 1063 hardly ever, or never? 7171 Subtotal 11341 Do you take part in sports or activities that are moderately energetic ….? {frequency in daily life} ... more than once a week, 6421 once a week, 1830 one to three times a month, 762 hardly ever, or never? 2326 Subtotal 11339 Do you take part in sports or activities that are mildly energetic….? {frequency in daily life} ... more than once a week, 8188 once a week, 1323 one to three times a month, 449 hardly ever, or never? 1381 Subtotal 11341 Table 1.9 shows partly 18% of the elderly in England take on vigorous physical exercises while majority prefer light and moderate exercises as shown in Table 1.9 Table 1.9 Summary of Physical Exercise Count Do you take part in sports or activities that are vigorous ….? {frequency in daily life} ... more than once a week, 2022 once a week, 1085 one to three times a month, 1063 hardly ever, or never? 7171 Subtotal 11341 Do you take part in sports or activities that are moderately energetic ….? {frequency in daily life} ... more than once a week, 6421 once a week, 1830 one to three times a month, 762 hardly ever, or never? 2326 Subtotal 11339 Do you take part in sports or activities that are mildly energetic….? {frequency in daily life} ... more than once a week, 8188 once a week, 1323 one to three times a month, 449 hardly ever, or never? 1381 Subtotal 11341 Further 80% of the aged citizens in England do moderately exercises as indicated from Table 1.9 and 88% of the old age still prefer mild exercises. 2. Health conditions variations i) The n2 = 0.2% indicating non-statistical significance of the results, further Table 2.4 shows F(1, 11520) = 22.9, p< .05 it indicates that F-test was significant. Thus the gender of the respondent does not have a relationship with the chronic ailments the person has. Table 2.4 Tests of Between-Subjects Effects Chronic ailments and sex Dependent Variable: Have/ Had Chronic Ailments Source Type III Sum of Squares df Mean Square F Sig. Partial Eta Squared Corrected Model 45.067a 1 45.067 22.949 .000 .002 Intercept 25328.759 1 25328.759 12897.748 .000 .528 dhsex 45.067 1 45.067 22.949 .000 .002 Error 22623.121 11520 1.964 Total 48413.000 11522 Corrected Total 22668.188 11521 a. R Squared = .002 (Adjusted R Squared = .002) ii) We conducted further tests to evaluate the health condition by marital status and one way ANOVA test was carried out. Table 2.1 and 2.2 shows that the F(1, 11514) = 41.8, p = 000. The Partial Eta Square .018 indicates a weak relationship between the health conditions and marital status. As F-test was significant we conducted a follow-up test taking care of Alpha error in the multivariate pairwise comparison. Thus p = .000,from the test variance of homogeneity, the test was significant (Betty R. Kirkwood, 2003) . Thus we can conclude that the overall health condition and marital status of the ELSA respondents has a weak significance and therefore the health of an individual does not depend on marital status as n2 accounts for 1.8% of the dependent variable total health conditions. Table 2.1 Levenes Test of Equality of Error Variancesa Dependent Variable: Have/ Had Chronic Ailments F df1 df2 Sig. 4.677 5 11514 .000 a. Design: Intercept + dimar Table 2.2 Tests of Between-Subjects Effects (Chronic ailments and marital status) Dependent Variable: Have/ Had Chronic Ailments Source Type III Sum of Squares df Mean Square F Sig. Partial Eta Squared Corrected Model 404.108a 5 80.822 41.806 .000 .018 Intercept 7574.070 1 7574.070 3917.761 .000 .254 dimar 404.108 5 80.822 41.806 .000 .018 Error 22259.609 11514 1.933 Total 48413.000 11520 Corrected Total 22663.718 11519 a. R Squared = .018 (Adjusted R Squared = .017) On conducting further parametric test, we found out that the mean age of 63 and 84 on age groups of 50-79 and above 80 respectively with the latter reporting less chronic ailments as shown in Table 2.3. Table 2.3 Group Statistics Age groups N Mean Std. Deviation Std. Error Mean Have/ had any chronic ailment 50 to 79 Years 10259 1.4326 1.38151 .01364 80 and above years 1263 2.0000 1.47025 .04137 Age variable from HH grid collapsed at 90 plus 50 to 79 Years 10259 62.76 8.275 .082 80 and above years 1263 84.49 4.941 .139 The above study was evaluated on independent t test and result as tabulated in Table 2.4, t(11520) = -13.6 p = .000 on equal variances; with 95 CI [-64874, -48607] indication a wide range in means. Thus the variance of chronic ailments variable is accounted for in ages of 50 -79 and above 80 years. The boxplot shows the cases of chronic ailments recorded on age groups in England. Table 2.5 shows that the majority of ELSA respondents are between age of 50 -79.. Table 2.5 Subject~s current legal marital status * Age groups Crosstabulation Count Age groups Total 50 to 79 Years 80 and above years Subject~s current legal marital status Single, that is never married 567 67 634 Married, first and only marriage 6050 366 6416 Remarried, second or later marriage 1186 69 1255 Legally separated 135 6 141 Divorced 1046 27 1073 Widowed 1273 728 2001 Total 10257 1263 11520 Table 2.6 Independent Samples Test Levenes Test for Equality of Variances t-test for Equality of Means F Sig. t df Sig. (2-tailed) Mean Difference Std. Error Difference 95% Confidence Interval of the Difference Lower Upper Prevalence Limitation of disability Equal variances assumed 594.800 .000 14.438 11520 .000 1.56669 .10851 1.35399 1.77938 Equal variances not assumed 11.041 1420.682 .000 1.56669 .14190 1.28834 1.84503 At 95% CI the equal variances indicated that the mean differrence was equal falling within CI as shown in Table 2.6.; t(11520) = 14.43, p < .01 indicating a significance thus the disability has a strong effects on the chronic ailments of the older citizens in England. 3.Health and income Further studies were carried out to find the relationship between income and health and whether they are correlated. Table 2.7 Tests of Between-Subjects Effects( Gender and Prevalence in disability) Dependent Variable: ASK OR CODE RESPONDENT~S SEX Source Type III Sum of Squares df Mean Square F Sig. Partial Eta Squared Corrected Model 5.133a 12 .428 1.727 .055 .002 Intercept 287.860 1 287.860 1162.084 .000 .092 limit_disa 5.133 12 .428 1.727 .055 .002 Error 2850.895 11509 .248 Total 30398.000 11522 Corrected Total 2856.028 11521 a. R Squared = .002 (Adjusted R Squared = .001) We conducted test to evaluate the relationship between the prevalence of disability and sex, we used 1 x 1 ANOVA to conduct the statistical test. The dependent variable sex has two levels female and male and the independent variable is continuous variable thus the results were not significant at 95% CI F(12, 11509) = 1.727, p = .055, the prevalence disability n2 = 0.2% indicating that the variance on dependent variable sex is negligible. From the findings we can conclude that the disability in old age has no relationship with the sex of the patient and but rather is based on individual patient. Table 3.1 Descriptive Statistics Mean Std. Deviation N Have/ Had Chronic Ailments 1.4972 1.40346 11417 BU equivalised total income 245.3813 281.30456 11417 Age variable from HH grid collapsed at 90 plus 65.1963 10.47889 11417 Table 3.1 shows the 50 and above mean income is £245.4 with an average mean of 65 years with a sample size N = 11417. Hypothesis Ho = Income of senior citizens aged 50 years and above in England is influenced by the health of the individual Ha = Income of senior citizens aged 50 years and above in England is not influenced by the health of the individual. Taking control of Alpha error we applied a Bonferroni approach to take care of the two correlations, and the results indicated p = 000 is nonsignificant. The variables Chronic ailments, Age, and income are multivariately distributed with normal attributes; thus the statistical relationship between the income and health can only be linear. The correlational analysis are presented in Table 3.2 the chronic ailments and the income are not statistically significant as ≤ .1. Thus r(11415) = -.103 , p = .000, indicates that there is no significant relationship between the variable income and chronic ailments. Thus Ho is rejected as the income of the older citizens is not correlated with the health of individual. Table 3.2 Correlations Control Variables Have/ Had Chronic Ailments BU equivalised total income Age variable from HH grid collapsed at 90 plus -none-a Have/ Had Chronic Ailments Correlation 1.000 -.103 .252 Significance (2-tailed) . .000 .000 Df 0 11415 11415 BU equivalised total income Correlation -.103 1.000 -.173 Significance (2-tailed) .000 . .000 Df 11415 0 11415 Age variable from HH grid collapsed at 90 plus Correlation .252 -.173 1.000 Significance (2-tailed) .000 .000 . Df 11415 11415 0 Age variable from HH grid collapsed at 90 plus Have/ Had Chronic Ailments Correlation 1.000 -.062 Significance (2-tailed) . .000 Df 0 11414 BU equivalised total income Correlation -.062 1.000 Significance (2-tailed) .000 . df 11414 0 The above graph shows there is no significant correlation in variables income and chronic ailments. A confounding variable marital status can bring up bias in the results as the income of the both spouses might have been combined to indicate that the respondent had a single income. The independent variable health and depended income. To remove the erroneous conclusion of the two variables we shall use statistical control by bootstrapping then stratifying the data before the analysis (Howell, 2013). Other variables that may be confounding variables are gender and HSE Social class. Gender can bring up the biasness in the results as income may be influenced by gender and social class; and may turn up do be depended of the two. 4. Multivariate regression Table 4.1 and 4.2 shows prediction of depression on disability 2 and disability 3. R2 = .16 , F= 535.1,p = .000, thus the disability 2 and disability 3 has an significant on depression. Table 4.1 Model R Square Adjusted R Square Std. Error of the Estimate F Sig. 1 .393a .155 .154 1.83992 535.141 .000b Model Summary R Table 4.2 Coefficientsa Model Unstandardized Coefficients t Sig. B Std. Error 1 (Constant) 1.490 .026 56.521 .000 Disability2 .605 .056 -10.817 .000 Disability3 1.381 .037 -10.354 .000 a. Dependent Variable: Depression score 0-8 zero is happiest Hence we get a model equation Depression = .605disability2 +1.38disabilty3 +1.49 The linear combination indicates disability has a great significance on the depression at 15.5% of disability, depression starts taking toll. From Table 4.3 and 4.4, R2 = .23, F = 658.8, p = 000, the results indicate that there is a great significance of disability2, disability3 and age on depression. Table 4.3 Model Summary Model R R Square Adjusted R Square Std. Error of the Estimate F Sig. 2 .478a .229 .228 2.5945 658.7501 .000b a. Predictors: (Constant), disability3, disability2, age, Table 4.4 Coefficientsa Model Unstandardized Coefficients t Sig. B Std. Error 1 (Constant) 1.490 .029 36.521 .000 disability2 -.005 .056 -0.017 .912 disability3 0.081 .077 -1.354 .405 Age 0.147 .010 13.914 .000 a. Dependent Variable: Depression score 0-8 zero is happiest Thus the prediction equation of the linear model will be; Depression = -0.005disability2 +0.081disability3 + 0.147age + 1.49 The linear equation shows a significance on depression, at 22.5% is much lower than the other model i.e 22.9 – 15.5 = 7.4%. The R2 indicates the goodness of fit in a multivariate regression. The double multivariate regression was further carried out to predict the overall depression. First analysis the predictors variables were disability2 and disability 3, and the second analysis age was added as a predictor besides the other two variables. The first analysis indicates depression appears to be predicted variables disability 2 and disability 3. On the second analysis three variables were tested. All the models they produced a significant relation to depression R2 = .16 and R2 = .23. From the results we can conclude depression, is significantly affected by the disability, but the bivariate correlation in age on the second analysis t(13.9), p < .05 indicates a age has a great correlation effects than other two predictors. Conclusion The depression in older age is much affected by the well being of the senior citizens being neglected and a lot of factor come into play. Thus, most ailments are reported between the age of 50 – 79 indicating more need to allocate more resources to this group. From the data, there is a need of citizens to treat the order generation with decorum and avoid isolating them as it can result to depression; thus inviting chronic ailments. As cognitive component contains beliefs, its prudent to involve the aged on physical exercise and social events to avoid loneliness and boredom thus averting depression (Dunn, 2015). The health conditions cut across the board in either gender and marital status an indication that chronic ailments and non-communicable diseases affect all regardless of sex, age, race or status. Thus improving old age social interaction, physical exercise is far much more effective and economical than resulting to expensive way of providing health services for the aged, thus improving the overall health age for senior citizens. Income plays a vital role healthy age society, thus older people are not social care cost, they can be engaged in other volunteer work delivering essential services to community and other senior citizens. Assumptions and limitation The data was collected is in a judgmental method from different health facilities around the country, and volunteer sampling was used as collection lasted for a period of time. The data is current in that it was collected recently. Limitations The data had incomplete ranges of data in different values and cells, in that some cases were missing from the sample. The data didn’t not explicit show the the quantitative attributes of the chronic diseases being treated or on treatment in various health facilities in England. The data does not indicate when the survey was taken or completed thus it can produce a bias information in regard to current trends. Coding of some data seems not be complete and some variables are missing the prerequisite details to give critical data analysis. Further studies need to be carried out to evaluate the level of productivity of the older people from 50 years and above. Ho = Can senior citizens aged 50 years and above contribute/ have an impact on economy. Ha = Senior citizens aged 50 years and above cannot contribute or have an impact on economy. Appendix 1 Syntax 1.1 Recode Chronic Ailments DATASET ACTIVATE DataSet1. RECODE hedia01 hedia02 hedia03 hedia04 hedia05 hedia06 hedia07 hedib01 hedib02 hedib03 hedib04 hedib05 hedib06 (-9=0) (-8=0) (-1=0) (96=0) (1=1) (2=1) (3=1) (4=1) (5=1) (6=1) (7=1) (8=1) (9=1) (95=1) INTO hec1 hec2 hec3 hec4 hec5 hec6 hec7 hec8 hec9 hec10 hec11 hec12 hec13. EXECUTE. COMPUTE Chronic_ail =sum(hec1 to hec13). 1.2 Recode for disabilities DATASET ACTIVATE DataSet1. RECODE scghqa scghqb scghqc scghqd scghqe scghqf scghqg scghqh scghqi scghqj scghqk scghql (-9=0) (-1=0) (1=1) (2=1) (3=1) (4=1) INTO Disability1 Disability2 Disability3 Disability4 Disability5 Disability6 Disability7 Disability8 Disability9 Disability10 Disability11 Disability12. VARIABLE LABELS Disability1 Disability1 /Disability2 Disability2 /Disability3 Disability3 /Disability4 Disability4 /Disability5 Disability5 /Disability6 Disability6 /Disability7 Disability7 /Disability8 Disability8 /Disability9 Disability9 /Disability10 Disability10 /Disability11 Disability11 /Disability12 Disability12. EXECUTE. COMPUTE limit_disa =sum(Disability1 to Disability12). EXECUTE. Works Cited Age UK, n.d. Healthy Ageing Evidence Review, London: s.n. Belinda Barton, J. P., 2014. Medical Statistics. 2nd ed. West Sussex: John Wiley & Sons. Betty R. Kirkwood, J. A. S., 2003. Essential Medical Statistics. 2nd ed. Massachusetts: Blackwell Science Ltd. Coolican, H., 2014. Research Methods and Statistics in Psychology. New York: Psychology Press. Crown copyright, 2014. Older people. [Online] Available at: https://www.gov.uk/government/policies/improving-opportunities-for-older-people [Accessed 31 March 2015]. Dunn, D. S., 2015. The Social Psychology of Disability. New York: Oxford University Press . Howell, D. C., 2013. Statistical Methods for Psychology. Belmont: Cengage Learning. Linda Seymour, E. G., 2004. Promoting mental health, London: s.n. Mary Kate Connolly, E. R., n.d. Dancing towards well-being in the Third Age, London: s.n. Triola, M. F., 2015. Essentials of Statistics. Boston: Pearson. Read More