CHAPTER 6 DATA ANALYSIS AND INTERPRETATION

208 CHAPTER 6 DATA ANALYSIS AND INTERPRETATION Sr. No. Content Page No. 6.1 Introduction 212 6.2 Reliability and Normality of Data 212 6.3 Descriptive Analysis 213 6.4 Cross Tabulation 218 6.5 Chi Square Test 223 6.5.1 Association between Gender of Respondents and 223 Method of Investment in Mutual Funds 6.5.2 Association between Age of Respondents and Method 224 of Investment in Mutual Funds 6.5.3 Association between Education Level of Respondents and Method of Investment in Mutual Funds 225 6.5.4 Association between Employment Status of 226 Respondents and Method of Investment in Mutual Funds 6.5.5 Association between Gender of Respondents and 227 Preference for Mutual Funds Type 6.5.6 Association between Age of Respondents and 228 Preference for Mutual Funds Type 6.5.7 Association between Education Level of Respondents 229 and Preference for Mutual Funds Type 6.5.8 Association between Employment Status of 229 Respondents and Preference for Mutual Funds Type

209 6.5.9 Association between Monthly Income of Respondents 230 and Preference for Mutual Funds Type 6.5.10 Association between Monthly Investment of 231 Respondents and Preference for Mutual Funds Type 6.5.11 Association between Gender of Respondents and Level 232 of Risk Preference 6.5.12 Association between Age of Respondents and Level of 232 Risk Preference 6.5.13 Association between Education Level of Respondents 233 and Level of Risk Preference 6.5.14 Association between Employment Status of 234 Respondents and Level of Risk Preference 6.5.15 Association between Monthly Income of Respondents 235 and Level of Risk Preference 6.5.16 Association between Monthly Investment of 236 Respondents and Level of Risk Preference 6.5.17 Association between Gender of Respondents and 236 Duration of Investment 6.5.18 Association between Age of Respondents and Duration 237 of Investment 6.5.19 Association between Employment Status of 238 Respondents and Duration of Investment 6.5.20 Association between Monthly Income of Respondents 239 and Duration of Investment 6.5.21 Association between Monthly Investment of 240 Respondents and Duration of Investment 6.6 Friedman ANOVA Test 241

210 6.6.1 Preference of Investors for Source of Information 241 about Mutual Funds 6.6.2 Preference of Investors for Different Mutual Funds 243 Schemes 6.6.3 Preference of Investors for Different Sponsors Mutual 245 Funds. 6.6.4 Preference of Investors for Different Investment 246 Avenues 6.7 Factor Analysis 249 6.8 Factor Analysis for Awareness About Mutual Funds 250 6.8.1 Bartlett Test of Spherity 250 6.8.2 Measures of Sampling Adequacy (MSA) 250 6.8.3 Anti-Image Correlation Matrix 251 6.8.4 Method of Factor Analysis 252 6.8.5 Method of Factor Rotation 252 6.8.6 Communalities 254 6.8.7 Eigenvalue and Total Variance Explained 264 6.8.8 Factor Loading 266 6.8.9 Revised Factor Extraction 268 6.8.10 Naming of Factors 270 6.9 ANOVA Analysis for Awareness 272 6.10 ANOVA Analysis for Awareness and Demographic Variables 273 6.10.1 ANOVA Analysis for Awareness * Gender 273 6.10.2 ANOVA Analysis for Awareness * Age 274 6.10.3 ANOVA Analysis for Awareness * Education Level 276 6.10.4 ANOVA Analysis for Awareness * Employment Status 279

211 6.11 Factor Analysis for Opportunities to Investment in Mutual Funds 282 6.11.1 Revised Factor Extraction 284 6.11.2 Naming of Factors 286 6.12 ANOVA Analysis for Opportunities and Demographic Variables 287 6.12.1 ANOVA Analysis for Opportunities * Gender 287 6.12.2 ANOVA Analysis for Opportunities * Age 288 6.12.3 ANOVA Analysis for Opportunities * Education Level 289 6.12.4 ANOVA Analysis for Opportunities * Employment 290 Status 6.13 Factor Analysis for Problems of Investing in Mutual Funds 295 6.13.1 Revised Factor Extraction 297 6.13.2 Naming of Factors 299 6.14 ANOVA Analysis for Problems and Demographic Variables 300 6.14.1 ANOVA Analysis for Problems * Gender 300 6.14.2 ANOVA Analysis for Problems * Age 301 6.14.3 ANOVA Analysis for Problems * Education Level 302 6.14.4 ANOVA Analysis for Problems * Employment Status 303 6.15 Summary 305

212 6.1 Introduction Once the raw data is collected, the next step is to analyze it to draw logical inferences from them. The data collected in a survey could be voluminous in nature, depending upon the size of the sample. It is necessary to arrange raw data collected during primary survey in useful fashion. The purpose of this chapter is to analyze and test the hypothesis created to achieve objective. To test the hypothesis analysis could be univariate, bivariate and multivariate in nature. In the univariate analysis one variable is analyzed at a time. In the bivariate analysis two variables are analysed together and examined for any possible association between them. In the multivariate analysis, the concern is to analyze more than two variables at a time. In present research various data analysis techniques have been used depending on the type of data and hypothesis framed. 6.2 Reliability and Normality of Data A multi item scale should be evaluated for accuracy and applicability (Malhotra, 2009). It is used to measure the strength of the scale. Questionnaire consist twenty two different statements on a 5-point scale. The content validity of the survey was assessed in a pre-test with 36 respondents not included in the sampling frame. Pre-test participants were asked to evaluate all aspects of the questionnaire, including the wording of individual items, the general flow and structure of the instrument and its comprehensiveness. Participants suggestions were then incorporated into the survey prior to its final use. The most widely used reliability measure is Cronbach s Alpha. So, in reliability analysis, the alpha (α) coefficient is calculated to find out the internal consistency of the items on the scale from 36 respondents. Number of questions Table 6.1 Reliability Statistics Cronbach s alpha Construct Cronbach s alpha value 63 Overall 0.932 22 Awareness 0.964 20 Opportunities 0.938 21 Problems 0.799

213 Alpha value of 0.6 or less generally indicates unsatisfactory level (Malhotra, 2009). But in this survey the overall and individual test value was found more than the required value that indicates good consistency among items and tools developed for study are reliable and hence researcher can proceed further. In case of normality, sample size has the effect of increasing statistical power by reducing sampling error Hair et. al. (2009), Larger sample size reduces the detrimental effects of non-normality. In small samples of 50 or less, significant departures from normality can have a substantial impact on the results but for sample size of 200 or more, however, the same effects may be negligible. Moreover, when group comparisons are made, such as ANOVA, the differing sample sizes between groups, if large enough, can even cancel out the detrimental effects. Thus, in most instances, as the sample sizes become large, researcher can be less concerned about non-normal variables. Normality can have serious effects in small samples (less than 50 cases), but the impact effectively diminishes when sample sizes reach 200 cases or more. Here in this research as sample size is 463, researcher has assumed that normality does exist. 6.3 Descriptive Analysis Descriptive analysis refers to transformation of raw data into a form that will facilitate easy understanding and interpretation. Descriptive analysis deals with summary measures relating to the sample data. The common ways to summarizing data are by calculating average, range, standard deviation, frequency and percentage distribution. The first thing to do when data analysis is taken up is to describe the sample. In following section characteristics of sample data gathered during primary survey are described.

214 Table 6.2 Descriptive Analysis for Demographic Variables Variable Range Frequency % Gender Male Female 348 115 75.2 24.8 Age 21-30 31-40 41-50 51-60 More than 60 195 132 88 41 7 42.1 28.5 19.0 8.9 1.5 Marital Status Single Married Others Education School Graduate Post graduate Professional Doctorate Others Family Size 1 person 2-3 persons 4-5 persons More than 5 persons Employment Status Salaried Businessman/Self Employed Housewife Student Retired Unemployed Type of House Stay Own Rented Investment in Yes Mutual Funds No Monthly Income Less than 10000 10001 to 15000 15001 to 20000 20001 to 25000 More than 25000 Monthly Investment in Mutual Funds Less than 5000 5001 to 10000 10001 to 15000 15001 to 20000 More than 20000 164 297 03 40 156 195 61 8 3 8 137 245 73 262 127 12 36 17 9 412 51 463 00 81 69 99 85 129 257 118 62 11 15 35.5 64.1 0.4 8.6 33.7 42.1 13.2 1.7 0.6 1.7 29.6 52.9 15.8 56.6 27.4 2.6 7.8 3.7 1.9 89.0 11.0 100 00 17.5 14.9 21.4 18.4 27.9 55.5 25.5 13.4 2.4 3.2

215 Preference for Method of Investment Preference for Mutual Funds type Preference for Risk Preference for Duration of Investment Online Offline Both Open Ended Close Ended Both High Risk Medium Risk Low Risk Less Than 1 Year 1 3 Years 3 5 Years 5 7 Years More Than 7 Years 101 197 165 146 96 221 65 249 149 53 174 139 57 40 21.8 42.6 35.6 31.6 20.7 47.7 14.0 53.8 32.2 11.5 37.6 30.0 12.3 8.6 Table 6.2 provides a summary of the demographic profile of the respondents. The result shows that a majority of the respondents were male (n = 348, 75.2 percent) with only 24.8 percent of the female respondent (n = 115). Age is represented by middle age respondents who represent almost 70 percent of the sample and which includes age group 21-30 and 31-40 (n=195 & 132) and those with age group between 41-50 and 51-60 represent 19 percent and 9 percent (n=88 & 41) respectively. 1.5 percent (n = 7) respondent were belong to more than 60 age group. Of the total sample, almost 64 percent (n=297) were married and 36 percent (n = 164) were belong to unmarried category. On the other hand, respondents with higher qualification such as professional and doctorate represent 13.2 and 1.7 (n=61 and 8), high educational level (postgraduates) represent 42 percent (n=195) of the sample, while medium education level (graduate) represent 33.7 percent (n=156) and lower education level (undergraduate and high school) represent 8.6 percent (n=40) of the sample. With respect to family size of sample, it can be seen that highest proportion of respondent n = 245 (53 percent) belong to moderate family size including 4-5 persons. About 30 percent respondent belong to family size including 2-3 persons in family (n=137). More than 5 persons in family members is represented by 16 percent (n = 73).

216 It is evident from table 6.2 with respect to employment status that highest proportion of the respondent were salaried employee and is represented by 56.6 percent (n=262) followed by businessman/self-employed 27.4 percent (n=127). Housewife, student, retired employee and unemployed respondent were represented by 12 percent (n=2.6), 7.8 percent (n=36), 3.7 percent (n=17) and 1.9 percent (n=9) respectively. Of the total respondents 89 percent (n=412) were staying in their own house while remaining 11 percent (n=51) were staying in rented house. While distributing respondents on the basis of monthly income, it is found that 17.5 percent of the respondents (n=81) earn less than Rs. 10000, 14.9 percent of respondents (n=69) earn between Rs. 10001 to Rs. 15000, 21.4 percent respondents (n=99) earn between Rs. 15001 to Rs. 20000 in a month, 18.4 percent of respondents (n=85) earn between Rs. 20001 to Rs. 25000 while 27.9 percent of the respondents (n=129) earn more than Rs. 25000 per month. Distribution of respondents based on monthly investment in mutual funds shows that 55.5 percent (n=257) respondents invest less than Rs. 5000 in mutual funds. 25.5 percent of respondents (n=118) invest between Rs. 5001 to Rs. 10000, 13.4 percent respondents (n=62) invest between Rs. 10001 to Rs. 15000, 2.4 percent respondents (n=11) invest between Rs. 15001 to Rs. 20000 and 3.2 percent of respondents (n=15) invest more than Rs. 20000 per month in mutual funds. It is found from table 6.2 that 21.8 percent of the respondents (n=101) prefer online mode of investment, 42.6 percent of respondents (n=197) prefer offline mode of investment while 165 respondents (35.6 percent) prefer both mode for investment in mutual funds. While distributing respondents based on type of scheme they prefer, it is found that 146 respondents (31.6 percent) prefer open ended schemes, 96 respondents (20.7 percent) respondents prefers close ended schemes and both schemes are preferred by 47.7 percent respondents (n=221).

217 High risk is preferred by 14 percent respondents (n=65), medium risk is preferred by 53.8 percent respondents while low risk is preferred by 149 respondents (32.2 percent). Table 6.2 also shows that, while distributing respondents based on preference of duration of investment 53 respondents (11.5 percent) prefer to invest for less than one year, 37.6 percent respondents (n=174) prefer to invest for 1-3 years, 30 percent of respondents (n=139) prefer to invest for 3-5 years, 12.3 percent of respondents (n=57) prefer to invest for 5-7 years while 40 respondents (8.6 percent) prefer to invest for more than 7 years. On the whole, sample respondents were more dominated by male, almost 90 percent belong to middle and upper middle age group; between 21-50, have higher levels of education (graduate and postgraduate), with moderate family size, and with 64 percent married (n=297) and 36 percent unmarried (n=164). Majority of the respondent 56.6 percent (n=262) were salaried employee followed by businessman/self-employed 27.4 percent (n=127). Of the total respondent almost 90 percent were having their own house and almost 30 percent of the respondents monthly income was more than 25000 while majority respondent; almost 80 percent (n= 375) were investing less than Rs. 10000 per month in mutual funds. Respondent tends to choose online and offline method of investment. Respondents are equality distributed as far as preference of method of investment is concern. For preference of scheme, higher proportion of respondent prefers to invest in both open ended and close ended schemes. It is also evident from the table 6.2 that majority of respondents (86 percent) are medium or low risk takers and higher proportion of respondents (67 percent) prefers to invest for short to medium term (1-5 years) in mutual funds.

218 6.4 Cross Tabulation Bivariate analysis examines the relationship between two variables. Different methods for analyzing bivariate analysis are available, of which cross tabulation is one of the important and frequently used methods. A cross tabulation counts the number of observations in each cross category of two variables. The descriptive result of a cross tabulation is a frequency count for each cell in the analysis. Following section deals with analysis of bivariate data using cross tabulation method. Table 6.3 Cross Tabulation for Gender * Age of Respondents Age Total 21-30 31-40 41-50 51-60 More than 60 Gender Male 132 98 76 35 7 348 Female 63 34 12 6 0 115 Total 195 132 88 41 7 463 Table 6.3 shows cross tabulation between gender and age of the respondents. It can be seen from table 6.3 that majority of the respondents were male and were belong to the age of 21 to 50 (88 percent), while in case of female majority of respondents belong to the age of 21 to 40 (85 percent). From the total respondents (n=463), 75 percent (n=348) were male and 25 percent (n=115) were female. Table 6.4 Cross Tabulation for Gender * Education of Respondents Education School Graduate Post- Graduate Professional Doctorate Others Total Gender Male 33 125 134 47 6 3 348 Female 7 31 61 14 2 0 115 Total 40 156 195 61 8 3 463 Cross tabulation of respondents based on gender and education status reveals that, from the total male respondents 38 percent (n=134) were post graduate, 36 percent

219 (n=125) were graduate and 14 percent respondents possess professional qualifications. Majority of female respondents 53 percent (n=61) were post graduate followed by graduate n=31 (27 percent) and professional qualification n=14 (12 percent). It can be concluded that from the total respondents, majority of them were literate and possess more than graduation qualification. Table 6.5 Cross Tabulation for Gender * Family Size of Respondents Family Size More than 1 Person 2-3 Persons 4-5 Persons 5 Persons Total Gender Male 8 100 187 53 348 Female 0 37 58 20 115 Total 8 137 245 73 463 Based on table 6.5 for cross tabulation, it is found that majority of male respondents (n=187) belong to moderate family size consisting 4-5 members and 53 male were belong to family size consisting more than 5 members. In case of female, higher proportion were belong to the family size of 4-5 members (n=58). Overall it can be said from table 6.5 that from the total respondents 68 percent (n=318), were belongs to the typical Indian moderate family size which consists more than 4 members. Table 6.6 Cross Tabulation for Gender * Employment Status of Respondents Employment Status Businessman/ Selfemployed Salaried Housewife Student Retired Unemployed Total Gender Male 177 115 1 31 17 7 348 Female 85 12 11 5 0 2 115 Total 262 127 12 36 17 9 463

220 While distributing respondents based on gender and their employment status, it was found that higher proportion in male and female both belongs to salaried employee category. In male category, almost 84 percent (n=292) were belong to salaried and businessman/self-employed category. Besides, almost 84 percent respondents (n=389) in male and female category both belong to salaried and businessman category. In all it can be said that majority of respondents belong to salaried and selfemployed category. Table 6.7 Cross Tabulation for Gender * Monthly Income of Respondents Monthly Income Less than 10001 to 15001 to 20001 to More than 10000 15000 20000 25000 25000 Total Gender Male 51 51 76 64 106 348 Female 30 18 23 21 23 115 Total 81 69 99 85 129 463 When classifying respondents based on monthly income with respect to their gender category, it was found that in male category respondents earning more than 25000 income per month were high in numbers (n=106), while other income group is almost equally distributed with their respective income. In female category there is a not vast difference between different income categories. Overall in male and female category both respondents who have earned more than 15000 in a month were high in numbers. Table 6.8 Cross Tabulation for Gender * Monthly Investment of Respondents Monthly Investment Less than 5001 to 10001 to 15001 to More than 5000 10000 15000 20000 20000 Total Gender Male 186 92 47 10 13 348 Female 71 26 15 1 2 115 Total 257 118 62 11 15 463

221 Though the majority respondents belongs to the income group of more than 25000 as seen in table no 6.7, the investment in mutual funds is less than 5000 in almost 50 percent cases. Further of the total male respondents almost 80 percent (n=278) respondents invest less than Rs. 10000 in mutual funds. In case of female also; almost 85 percent (n=97) invest less than Rs. 10000 in mutual funds. Table 6.9 Cross Tabulation for Age * Employment Status of Respondents Employment Status Businessman/ Self- Salaried employed Housewife Student Retired Unemployed Total Age 21-30 128 22 4 34 0 7 195 31-40 76 49 4 1 1 1 132 41-50 36 45 3 1 2 1 88 51-60 22 10 1 0 8 0 41 More than 60 0 1 0 0 6 0 7 Total 262 127 12 36 17 9 463 From the table 6.9 it can be seen that high proportion of respondents consists salaried employee. Further, almost 50 percent of salaried employee belong to the age category of 21-30 years, while remaining 50 percent salaried employee belong to the age category between 31-60 years. Second category employees are businessman which includes 27 percent of respondent. The portion of housewife, student, retired employee and unemployed category includes less than 10 percent respondent in each category and their portion in each category is very nominal.

222 Table 6.10 - Cross Tabulation for Age * Monthly Income of Respondents Less than 10000 10001 to 15000 Monthly Income 15001 to 20000 20001 to 25000 More than 25000 Total Age 21-30 61 38 47 24 25 195 31-40 9 14 24 33 52 132 41-50 7 9 20 15 37 88 51-60 2 6 6 12 15 41 More than 60 2 2 2 1 0 7 Total 81 69 99 85 129 463 Cross tabulation of monthly income with age reveals that 61 respondents earn less than 10000 in a month belong to the age category of 21-30. In the same age category respondents who belong 10001 to 15000, 15001 to 20000, 20001 to 25000 and more than 25000 income groups were 38, 47, 24 and 25 respectively in numbers. In the age category of 31-40 the respondents whose income was less than 10000, 10001 to 15000, 15001 to 20000, 20001 to 25000 and more than 25000 income group were 9, 14, 24, 33 and 52 respectively in numbers. Overall it can be said that majority of respondents were belong to the age group of 21-40. Table 6.11 Cross Tabulation for Age * Monthly Investment of Respondents Less than 5000 5001 to 10000 Monthly Investment 10001 to 15000 15001 to 20000 More than 20000 Total Age 21-30 136 39 15 2 3 195 31-40 60 43 20 1 8 132 41-50 37 23 20 6 2 88 51-60 19 11 7 2 2 41 More than 60 5 2 0 0 0 7 Total 257 118 62 11 15 463

223 Though the income of majority of respondents was more than 15000 per month; the investment in mutual funds is less than 5000 in 55 percent cases. Further, those who belong to age group between 21-30, invest less than 5000 in 70 percent cases (n=136). Respondents in the age group of 21-30 consist 136, 39, 15, 2 and 3 respondents who invest less than 5000, 5001 to 10000, 10001 to 15000, 15001 to 20000 and more than 20000 in a month respectively. In the age group of 31-40, total 132 respondents are included. From the total respondents in the same group 60 respondents invest less than 5000, 43 invest between 5001 to 10000, 20 invest 10001 to 15000, 01 invest 15001 to 20000 and 8 respondents invest more than 20000 in a month in mutual funds. The proportion of age group 41-50, 51-60 and more than 60 includes 88, 41 and 7 respondents respectively who belong to different investment category. 6.5 Chi Square Test The purpose of chi-square test is to show the relationship or lack of relationship between two variables. It is used to test the statistical significance of the observed association in a cross tabulation (Malhotra, 2009). It assists in determining whether a systematic association exists between the two variables. The test is conducted by computing the cell frequencies that would be expected if no association were present between the variables, given the existing row and column totals. A number of tests are available to determine if the relationship between two cross-tabulated variables is significant. One of the common tests is chi-square test. In present research to study the association between two variables chi-square statistics has been used. For below; all hypotheses test were performed at 5 percent level of significance. 6.5.1 Association between Gender of Respondents and Method of Investment in Mutual Funds Hypothesis H 0 H 1 There is no significant association between gender of respondents and method of investment in mutual funds. There is significant association between gender of respondents and method of investment in mutual funds.

224 Table 6.12 Association between Gender of Respondents * Method of Investment Chi-Square Tests Value Df Sig. Pearson Chi-Square 3.928 2.140 Likelihood Ratio 3.904 2.142 Linear-by-Linear Association.017 1.897 N of Valid Cases 463 Table 6.12 present the output of chi-square test. The person chi square value is 0.140 at 2 degrees of freedom, which is more than cut of value 0.05 at 95 percent confidence level. Therefore, null hypothesis is accepted and hence it can be said that there is no association between gender of the respondents and their method of investment. It can be also said that gender and method of investment are independent. 6.5.2 Association between Age of Respondents and Method of Investment in Mutual Funds Hypotheses H 0 H 1 There is no significant association between age of respondents and method of investment in mutual funds. There is significant association between age of respondents and method of investment in mutual funds. Table 6.13 Association between Age of Respondents * Method of Investment Chi-Square Tests Value Df Sig. Pearson Chi-Square 12.076 8.148 Likelihood Ratio 13.276 8.103 Linear-by-Linear Association.462 1.497 N of Valid Cases 463

225 Table 6.13 present the output of chi-square test. The Pearson Chi Square value is 0.148 at 8 degrees of freedom, which is more than cut of value 0.05 at 95 percent confidence level. Therefore, null hypothesis accepted and hence it can be said that there is no association between age of the respondents and their method of investment. It can be also said that age and method of investment, both are independent of each other and are not significantly related. 6.5.3 Association between Education Level of Respondents and Method of Investment in Mutual Funds Hypotheses H 0 H 1 There is no significant association between education level of respondents and method of investment in mutual funds. There is significant association between education level of respondents and method of investment in mutual funds. Table 6.14 Association between Education Level of Respondents * Method of Investment Chi-Square Tests Value df Sig. Pearson Chi-Square 8.084 10.621 Likelihood Ratio 9.223 10.511 Linear-by-Linear Association.024 1.877 N of Valid Cases 463 Table 6.14 present the output of chi-square test. The Pearson Chi Square value is 0.621 at 10 degrees of freedom, which is more than cut of value 0.05 at 95 percent confidence level. Therefore, null hypothesis accepted and hence it can be said that there is no association between education level of the respondents and their method of investment. It can be also said that education level and method of investment are independent.

226 6.5.4 Association between Employment Status of Respondents and Method of Investment in Mutual Funds Hypotheses H 0 H 1 There is no significant association between employment status of respondents and method of investment in mutual funds. There is significant association between employment status of respondents and method of investment in mutual funds. Table 6.15 Association between Employment Status of Respondents * Method of Investment Chi-Square Tests Value df Sig. Pearson Chi-Square 30.559 10.001 Likelihood Ratio 31.250 10.001 Linear-by-Linear Association 2.082 1.149 N of Valid Cases 463 Table 6.16 Cramer s V for Employment Status of Respondents * Method of Investment Value Sig. Nominal by Phi.257.001 Nominal Cramer s V.182.001 Contingency Coefficient.249.001 N of Valid Cases 463 Table 6.15 present the output of chi-square test. The Pearson Chi Square value is 0.001 at 10 degrees of freedom, which is less than cut of value 0.05 at 95 percent confidence level. Therefore, null hypothesis is rejected and hence it can be said that there is an association between employment status of the respondents and their method of investment. It can be also said that employment status and method of investment are significantly related.

227 Though in case of employment status and method of investment association is statistically significant, the chi square test does not tell the strength or degree of association. Strength of association or degree of association is of interest when the association is statistically significant. The strength of association can be measured by phi correlation coefficient, Cramers V, Contingency coefficient and the Lamda Coefficient (Malhotra, 2009). It is used based on the properties of the variables. As the row and column size are more than 2*2 Cramer s V value has been considered for further interpretation. The Cramer s V value varies between 0 to +1. If it takes the value of 0 when there is no association while +1 shows perfect positive association. A large value of V merely indicates a high degree of association, but does not indicate how the variables are associated. Here in this case the value for Cramer s V is 18.2 (.182) which shows that the association is not very strong. 6.5.5 Association between Gender of Respondents and Type of Mutual Funds Preferred Hypotheses H 0 H 1 There is no significant association between gender of the respondents and type of mutual funds preferred. There is significant association between gender of the respondents and type of mutual funds preferred. Table 6.17 Association between Gender of Respondents * Preference for Mutual Funds Type Chi-Square Tests Value df Sig. Pearson Chi-Square.060 2.970 Likelihood Ratio.060 2.970 Linear-by-Linear Association.006 1.939 N of Valid Cases 463 Table 6.17 present the output of chi-square test. The Pearson Chi Square value is 0.970 at 2 degrees of freedom, which is more than cut of value 0.05 at 95 percent

228 confidence level. Therefore, null hypothesis accepted and hence it can be said that there is no association between gender of the respondents and preference for mutual funds type. It can be also said that gender and preference of mutual funds type are independent. 6.5.6 Association between Age of Respondents and Type of Mutual Funds Preferred Hypotheses H 0 H 1 There is no significant association between age of the respondents and type of mutual funds preferred. There is significant association between age of the respondents and type of mutual funds preferred. Table 6.18 Association between Age of Respondents * Preference for Mutual Funds Type Chi-Square Tests Value df Sig. Pearson Chi-Square 7.000 8.537 Likelihood Ratio 8.120 8.422 Linear-by-Linear Association.548 1.459 N of Valid Cases 463 Table 6.18 present the output of chi-square test. The Pearson Chi Square value is 0.537 at 8 degrees of freedom, which is more than cut of value 0.05 at 95 percent confidence level. Therefore, null hypothesis accepted and hence it can be said that there is no association between age of the respondents and preference for mutual funds type. It can be also said that age and preference for mutual funds type are independent.

229 6.5.7 Association between Education Level of Respondents and Type of Mutual Funds Preferred Hypotheses H 0 H 1 There is no significant association between education level of the respondents and type of mutual funds preferred. There is significant association between education level of the respondents and type of mutual funds preferred. Table 6.19 Association between Education Level of Respondents * Preference for Mutual Funds Type Chi-Square Tests Value df Sig. Pearson Chi-Square 6.720 10.752 Likelihood Ratio 7.018 10.724 Linear-by-Linear Association 1.888 1.169 N of Valid Cases 463 Table 6.19 present the output of chi-square test. The Pearson Chi Square value is 0.752 at 10 degrees of freedom, which is more than cut of value 0.05 at 95 percent confidence level. Therefore, null hypothesis accepted and hence it can be said that there is no association between education level of the respondents and preference for mutual funds type. It can be also said that education level and preference for mutual funds type are independent. 6.5.8 Association between Employment Status of Respondents and Type of Mutual Funds Preferred Hypotheses H 0 H 1 There is no significant association between employment status of the respondents and type of mutual funds preferred. There is significant association between employment status of the respondents and type of mutual funds preferred.

230 Table 6.20 Association between Employment Status of Respondents * Preference for Mutual Funds Type Chi-Square Tests Value Df Sig. Pearson Chi-Square 16.502 10.086 Likelihood Ratio 17.380 10.066 Linear-by-Linear Association 1.437 1.231 N of Valid Cases 463 Table 6.20 present the output of chi-square test. The Pearson Chi Square value is 0.086 at 10 degrees of freedom, which is more than cut of value 0.05 at 95 percent confidence level. Therefore, null hypothesis accepted and hence it can be said that there is no association between employment status of the respondents and preference for mutual funds type. It can be also said that employment status and preference for mutual funds type are independent. 6.5.9 Association between Monthly Income of Respondents and Type of Mutual Funds Preferred Hypotheses H 0 H 1 There is no significant association between monthly income of the respondents and type of mutual funds preferred. There is significant association between monthly income of the respondents and type of mutual funds preferred. Table 6.21 Association between Monthly Income of Respondents * Preference for Mutual Funds Type Chi-Square Tests Value Df Sig. Pearson Chi-Square 13.934 8.083 Likelihood Ratio 14.578 8.068 Linear-by-Linear Association.233 1.630 N of Valid Cases 463

231 Table 6.21 present the output of chi-square test. The Pearson Chi Square value is 0.083 at 8 degrees of freedom, which is more than cut of value 0.05 at 95 percent confidence level. Therefore, null hypothesis accepted and hence it can be said that there is no association between monthly income of the respondents and preference for mutual funds type. It can be also said that monthly income and preference for mutual funds type are independent. 6.5.10 Association between Monthly Investment of Respondents and Type of Mutual Funds Preferred Hypotheses H 0 H 1 There is no significant association between monthly investment of the respondents and type of mutual funds preferred. There is significant association between monthly investment of the respondents and type of mutual funds preferred. Table 6.22 Association between Monthly Investment of Respondents * Preference for Mutual Funds Type Chi-Square Tests Value Df Sig. Pearson Chi-Square 9.807 8.279 Likelihood Ratio 9.569 8.297 Linear-by-Linear Association.242 1.623 N of Valid Cases 463 Table 6.22 present the output of chi-square test. The Pearson Chi Square value is 0.279 at 8 degrees of freedom, which is more than cut of value 0.05 at 95 percent confidence level. Therefore, null hypothesis accepted and hence it can be said that there is no association between monthly investment of the respondents and preference for mutual funds type. It can be also said that monthly investment and preference for mutual funds type are independent.

232 6.5.11 Association between Gender of Respondents and Level of Risk Preferred Hypotheses H 0 There is no significant association between gender of the respondents and level of risk preferred. H 1 There is significant association between gender of the respondents and level of risk preferred. Table 6.23 Association between Gender of Respondents * Preference for Risk Chi-Square Tests Value Df Sig. Pearson Chi-Square 2.327 2.312 Likelihood Ratio 2.397 2.302 Linear-by-Linear Association 2.245 1.134 N of Valid Cases 463 Table 6.23 present the output of chi-square test. The Pearson Chi Square value is 0.312 at 2 degrees of freedom, which is more than cut of value 0.05 at 95 percent confidence level. Therefore, null hypothesis accepted and hence it can be said that there is no association between gender of the respondents and level of risk preferred. It can be also said that gender and preference for risk level are independent. 6.5.12 Association between Age of Respondents and Level of Risk Preferred Hypotheses H 0 H 1 There is no significant association between age of the respondents and level of risk preferred. There is significant association between age of the respondents and level of risk preferred.

233 Table 6.24 Association between Age of Respondents * Preference for Risk Chi-Square Tests Value Df Sig. Pearson Chi-Square 13.584 8.093 Likelihood Ratio 12.973 8.113 Linear-by-Linear Association 3.761 1.052 N of Valid Cases 463 Table 6.24 present the output of chi-square test. The Pearson Chi Square value is 0.093 at 8 degrees of freedom, which is more than cut of value 0.05 at 95 percent confidence level. Therefore, null hypothesis accepted and hence it can be said that there is no association between age of the respondents and level of risk preferred. It can be also said that age and preference for risk level are independent. 6.5.13 Association between Education Level of Respondents and Level of Risk Preferred Hypotheses H 0 H 1 There is no significant association between education level of the respondents and level of risk preferred. There is significant association between education level of the respondents and level of risk preferred. Table 6.25 Association between Education Level of Respondents * Preference for Risk Chi-Square Tests Value Df Sig. Pearson Chi-Square 18.025 10.055 Likelihood Ratio 18.764 10.043 Linear-by-Linear Association 4.632 1.031 N of Valid Cases 463

234 Table 6.25 present the output of chi-square test. The Pearson Chi Square value is 0.055 at 10 degrees of freedom, which is more than cut of value 0.05 at 95 percent confidence level. Therefore, null hypothesis accepted and hence it can be said that there is no association between education level of the respondents and level of risk preferred. It can be also said that education level and preference for risk level are independent. 6.5.14 Association between Employment Status of Respondents and Level of Risk Preferred Hypotheses H 0 H 1 There is no significant association between employment status of the respondents and level of risk preferred. There is significant association between employment status of the respondents and level of risk preferred. Table 6.26 Association between Employment Status of Respondents * Preference for Risk Chi-Square Tests Value Df Sig. Pearson Chi-Square 15.930 10.102 Likelihood Ratio 15.347 10.120 Linear-by-Linear Association 6.606 1.010 N of Valid Cases 463 Table 6.26 present the output of chi-square test. The Pearson Chi Square value is 0.102 at 10 degrees of freedom, which is more than cut of value 0.05 at 95 percent confidence level. Therefore, null hypothesis accepted and hence it can be said that there is no association between employment status of the respondents and level of risk preferred. It can be also said that employment status and preference for risk level are independent.

235 6.5.15 Association between Monthly Income of Respondents and Level of Risk Preferred Hypotheses H 0 There is no significant association between monthly income of the respondents and level of risk preferred. H 1 There is significant association between monthly income of the respondents and level of risk preferred. Table 6.27 Association between Monthly Income * Preference for Risk Chi-Square Tests Value df Sig. Pearson Chi-Square 23.248 8.003 Likelihood Ratio 24.446 8.002 Linear-by-Linear Association 4.104 1.043 N of Valid Cases 463 Symmetric Measures Value Sig. Nominal by Nominal Phi.224.003 Cramer s V.158.003 Contingency Coefficient.219.003 N of Valid Cases 463 Table 6.27 present the output of chi-square test. The Pearson Chi Square value is 0.003 at 8 degrees of freedom, which is less than cut of value 0.05 at 95 percent confidence level. Therefore, null hypothesis rejected and hence it can be said that there is association between monthly income of the respondents and level of risk preferred. It can be also said that monthly income and preference for risk level are significantly related. As the row and column size are more than 2*2, Cramer s V value has been consider to measure the degree or strength of association for further interpretation. Here in this case the value for Cramer s V is 15.8 (.158) which shows that the association is not very strong.

236 6.5.16 Association between Monthly Investment of Respondents and Level of Risk Preferred Hypotheses H 0 H 1 There is no significant association between monthly investment of the respondents and level of risk preferred. There is significant association between monthly investment of the respondents and level of risk preferred. Table 6.28 Association between Monthly Investment of Respondents * Preference for Risk Chi-Square Tests Value Df Sig. Pearson Chi-Square 11.313 8.185 Likelihood Ratio 11.674 8.166 Linear-by-Linear Association 2.841 1.092 N of Valid Cases 463 Table 6.28 present the output of chi-square test. The Pearson Chi Square value is 0.185 at 8 degrees of freedom, which is more than cut of value 0.05 at 95 percent confidence level. Therefore, null hypothesis accepted and hence it can be said that there is no association between monthly investment of the respondents and level of risk preferred. It can be also said that monthly investment and preference for risk level are independent. 6.5.17 Association between Gender of Respondents and Duration of Investment Hypotheses H 0 H 1 There is no significant association between gender of the respondents and duration of investment in mutual funds. There is significant association between gender of the respondents and duration of investment in mutual funds.

237 Table 6.29 Association between Gender of Respondents * Duration of Investment Chi-Square Tests Value Df Sig. Pearson Chi-Square 8.484 4.075 Likelihood Ratio 9.328 4.053 Linear-by-Linear Association 1.738 1.187 N of Valid Cases 463 Table 6.29 present the output of chi-square test. The Pearson Chi Square value is 0.075 at 4 degrees of freedom, which is more than cut of value 0.05 at 95 percent confidence level. Therefore, null hypothesis accepted and hence it can be said that there is no association between gender of the respondents and preference for duration of investment in mutual funds. It can be also said that gender and investment duration preference are independent. 6.5.18 Association between Age of Respondents and Duration of Investment Hypotheses H 0 There is no significant association between age of the respondents and duration of investment in mutual funds. H 1 There is significant association between age of the respondents and duration of investment in mutual funds. Table 6.30 Association between Age of Respondents * Duration of Investment Chi-Square Tests Value df Sig. Pearson Chi-Square 41.019 16.001 Likelihood Ratio 41.213 16.001 Linear-by-Linear Association 26.879 1.000 N of Valid Cases 463 Symmetric Measures Value Sig. Nominal by Nominal Phi.298.001 Cramer s V.149.001 Contingency Coefficient.285.001 N of Valid Cases 463

238 Table 6.30 present the output of chi-square test. The Pearson Chi Square value is 0.001 at 16 degrees of freedom, which is less than cut of value 0.05 at 95 percent confidence level. Therefore, null hypothesis rejected and hence it can be said that there is association between age of the respondents and duration of investment in mutual funds. It can be also said that age and duration preference are significantly related. As the row and column size are more than 2*2, Cramer s V value has been consider to measure the degree or strength of association for further interpretation. Here in this case the value for Cramer s V is 14.9 (.149) which shows that the association is not very strong. 6.5.19 Association between Employment Status of Respondents and Duration of Investment Hypotheses H 0 There is no significant association between employment status of the respondents and duration of investment in mutual funds. H 1 There is significant association between employment status of the respondents and duration of investment in mutual funds. Table 6.31 Association between Employment Status of Respondents * Duration of Investment Chi-Square Tests Value df Sig. Pearson Chi-Square 39.831 20.005 Likelihood Ratio 38.258 20.008 Linear-by-Linear Association.689 1.406 N of Valid Cases 463 Symmetric Measures Value Sig. Nominal by Nominal Phi.293.005 Cramer s V.147.005 Contingency Coefficient.281.005 N of Valid Cases 463

239 Table 6.31 present the output of chi-square test. The Pearson Chi Square value is 0.005 at 20 degrees of freedom, which is equal to the cut of value 0.05 at 95 percent confidence level. Therefore, null hypothesis rejected and hence it can be said that there is association between employment status of the respondents and duration of investment in mutual funds. It can be also said that employment status and duration preference are significantly related. As the row and column size are more than 2*2, Cramer s V value has been consider to measure the degree or strength of association for further interpretation. Here in this case the value for Cramer s V is 14.7 (.147) which shows that the association is not very strong. 6.5.20 Association between Monthly Income of Respondents and Duration of Hypotheses H 0 H 1 Investment There is no significant association between monthly income of the respondents and duration of investment in mutual funds. There is significant association between monthly income of the respondents and duration of investment in mutual funds. Table 6.32 Association between Monthly Income of Respondents * Duration of Investment Chi-Square Tests Value Df Sig. Pearson Chi-Square 51.630 16.000 Likelihood Ratio 51.170 16.000 Linear-by-Linear Association 24.591 1.000 N of Valid Cases 463 Symmetric Measures Value Sig. Nominal by Nominal Phi.334.000 Cramer s V.167.000 Contingency Coefficient.317.000 N of Valid Cases 463 Table 6.32 present the output of chi-square test. The Pearson Chi Square value is 0.000 at 16 degrees of freedom, which is less than cut of value 0.05 at 95 percent confidence level. Therefore, null hypothesis rejected and hence it can be said that

240 there is association between monthly income of the respondents and duration of investment in mutual funds. It can be also said that monthly income and duration preference are significantly related. As the row and column size are more than 2*2, Cramer s V value has been consider to measure the degree or strength of association for further interpretation. Here in this case the value for Cramer s V is 16.7 (.167) which shows that the association is not very strong. 6.5.21 Association between Monthly Investment of Respondents and Duration of Hypotheses H 0 H 1 Investment There is no significant association between monthly investment of the respondents and duration of investment in mutual funds. There is significant association between monthly investment of the respondents and duration of investment in mutual funds. Table 6.33 Association between Monthly Investment of Respondents * Duration of Investment Chi-Square Tests Value df Sig. Pearson Chi-Square 41.307 16.001 Likelihood Ratio 43.257 16.000 Linear-by-Linear Association 22.604 1.000 N of Valid Cases 463 Symmetric Measures Value Sig. Nominal by Nominal Phi.299.001 Cramer s V.149.001 Contingency Coefficient.286.001 N of Valid Cases 463 Table 6.33 present the output of chi-square test. The Pearson Chi Square value is 0.001 at 16 degrees of freedom, which is less than the cut of value 0.05 at 95 percent confidence level. Therefore, null hypothesis rejected and hence it can be said that there is association between monthly investment of the respondents and duration of

241 investment in mutual funds. It can be also said that monthly investment and duration preference are significantly related. As the row and column size are more than 2*2, Cramer s V value has been consider to measure the degree or strength of association for further interpretation. Here in this case the value for Cramer s V is 14.9 (.149) which shows that the association is not very strong. 6.6 Friedman ANOVA Test Friedman test is a nonparametric version of the randomized block design ANOVA. It is used when the observations are more than paired; such each block or person is assigned to all treatments (Malhotra, 2009). Since the Friedman test is based on ranks, it is especially useful for testing treatment effects when the observations are in the form of ranks; where ANOVA test cannot be used. When the assumption of normality may not hold, Friedman test instead of the parametric ANOVA test is used. 6.6.1 Preference of Respondents for Source of Information about Mutual Funds Hypotheses H0 H1 There is no significant difference in mean score of preferences given by respondents for different source of information about mutual funds. There is significant difference in mean score of preferences given by respondents for different source of information about mutual funds. Table 6.34 Cross Tabulation for Source of Information * Gender of Respondents Gender Source of information Male Female Total Newspaper Advertisement 52 17 69 Agent 51 21 72 Bank 36 13 49 Financial Advisor 67 24 91 Stock Broker 40 11 51

242 Internet 38 8 46 Friends 33 8 41 Relatives 11 5 16 Television 20 8 28 Count 348 115 463 Table 6.34 reveals that from the total sources of information available to retail investor first preference given by both male and female respondents (n=91) is financial advisor. In male category, second preference was given to newspaper advertisement (n=52) followed by agent, stock broker, internet, bank, friends, television and relatives respectively. While in female category second preference is given to agent (n=21) followed by newspaper advertisement bank, stock broker and relatives. There are equal preferences of female category in case of internet, friends and television. In both male and female categories; first preference was given to financial advisor while seeking information about mutual funds. Table 6.35 Mean Rank for Preference for Source of Information Source of Information Mean Rank Newspaper Advertisement 4.92 Agent 4.26 Bank 4.62 Financial Advisor 4.17 Stock Broker 4.80 Internet 5.19 Friends 5.15 Relatives 6.02 Television 5.86 Table 6.36 Friedman Test for Preference of Source of Information N 463 Chi-Square 201.437 Df 8 Sig..000

243 Table 6.35 and 6.36 presents the output of Friedman ANOVA test. The Friedman test value is 0.000 at 8 degrees of freedom, which is less than cut of value 0.05 at 95 percent confidence level. Therefore, null hypothesis is rejected and hence it can be said that there is significance difference in mean score of preferences given by male and female. Further from mean value analysis it is indicative that financial advisor (mean rank = 4.17) is the most important and preferred source for collecting information regarding mutual funds from the respondent view point followed by Agent (mean rank = 4.26). The mean rank (6.02) of the source relatives is highest, so it can be said that relatives are the least reliable and preferred source of information from the respondent point of view. 6.6.2 Preference of Respondents for Different Mutual Funds Schemes Hypotheses H0 H1 There is no significant difference in mean score of preferences given by respondents for different schemes of mutual funds. There is significant difference in mean score of preferences given by respondents for different schemes of mutual funds. Table 6.37 Cross Tabulation for Preference of Schemes * Education Level of Different mutual funds schemes Regular / Debt Schemes Growth Schemes Balanced Schemes School Graduate Respondents Education Post- Graduate Professional Doctorate Others Total 7 26 49 16 2 2 102 13 66 61 15 3 1 159 8 14 33 9 1 0 65 Liquid Schemes 4 22 23 5 0 0 54 Tax Savings Schemes 8 28 30 16 2 0 84 Total 40 156 195 61 8 3 463