FREE PROJECT ON THE EFFECT OF MORTALITY RATE ON POPULATION GROWTH IN NIGERIA

FREE PROJECT ON THE EFFECT OF MORTALITY RATE ON POPULATION GROWTH IN NIGERIA




ABSTRACT


This research work was carried out to study the effect of mortality rate on population growth in Nigeria. It was aimed at determine whether the deaths depends on age by studying the relationship between the death rate and age and to determine whether there is significance difference between male and female deaths in Nigeria. Secondary data were used, it’s extracted from National Population Commission, NDHS on mortality by age specific death rate and gender, and it covers the period: 2002-2006. The statistical techniques used for data analysis are Chi-square test of independence, Correlation Coefficient and T-test. From the analysis, it was found that both the gender and the age contributed to the deaths. The differences between male and female deaths are due to chance variation which implies that there is significance between the male and female deaths and the means of age groups do differs significantly.















TABLE OF CONTENT

Title page……………………………………………………………………………………….....i

Certification………………………………………………………………………………………ii

Dedication……………………………………………………………………………………….iii

Acknowledgement………………………………………………………………………………iv

Abstract…………………………………………………………………………….....................v

CHAPTER ONE: INTRODUCTION

Introduction………………………………………………………………………………..1

1.1 Statement of Problem………………………………………………………...........................3

1.2 Aim and Objectives…………………………………………………………………………..4

1.3 Significance of the Study…………………………………………………………………….4

1.4 Scope and Limitation of the Study…………………………………………...........................4

1.5 Definition of Terms…………………………………………………………………………...5

CHAPTER TWO: LITERATURE REVIEW

Literature Review…………………………………………………………………………..6

CHAPTER THREE: RESEARCH METHODOLOGY

3.0 Research Methodology……………………………………………………………………...14

3.1 Data Collection Methods……………………………………………………………………14

3.2 Types of Data………………………………………………………………………………..14

3.3 Sources of Data……………………………………………………………………………...14 

3.4 Method of Analysis………………………………………………………………………….15

       3.4.1 Contingency Table…..………………………………………………………………...15

       3.4.2 Chi – Squared Test of Independence….……………………………………………...15

       3.4.3 Properties of Chi-Square.……………………………………………………………..16

       3.4.4 Five Step of Chi-Square Test of Independence……………….……………………...17

       3.4.5 Tests of Hypotheses for Chi-Square Test of Independence…..………………………17

       3.4.6 Correlation Coefficient………………………………………………………………..18

       3.4.7 Tests of Hypotheses for Correlation Coefficient……………………………………...19

       3.4.8 T-Test………………………………………………………………………………….19

       3.4.9 Tests of Hypotheses for T-Test……….……………………………………………….20

CHAPTER FOUR: DATA PRESENTATION AND ANALYSIS

4.0 Data Presentation and Analysis …………………………………………………………….21

4.1Data Presentation ………………………………………………………………………..…..21

4.2 Data Analysis …….………..………………………………………………………………..22

CHAPTER FIVE: SUMMARY, CONCLUSION AND RECOMENDATION

5.0 Summary, Conclusion and Recommendation………………………………………………30

5.1 Summary ……………………………………………………………………………………30

5.2 Conclusion ……………………………………………………………….…........................31

5.3 Recommendations…………………………………………………………..........................31

References………………………………………………………................................................32


















CHAPTER ONE

1.0    INTRODUCTION

Mortality rate is the number of deaths in a given area or period, or from a particular cause. It is one of the factors affecting the population of developing countries of the world. 

Mortality rate is a measure of the number of deaths (in general, or due to a specific cause) in a population, scaled to the size of that population, per unit of time. Mortality rate is typically expressed in units of deaths per 1,000 individuals per year; thus, a mortality rate of 9.5 (out of 1,000) in a population of 1,000 would mean 9.5 deaths per year in that entire population, or 0.95% out of the total. 

The crude death rate is the total number of deaths per year per 1,000 people. The crude death rate depends on the age (and gender) specific mortality rates and the age (and gender) distribution of the population. The number of deaths per 1,000 people can be higher in developed nations than in less-developed countries, despite a higher life expectancy in developed countries due to better standards of health. This happens because developed countries typically have a much higher proportion of older people, due to both lower birth rate and lower mortality rates. A more complete picture of mortality is given by a life table, which shows the mortality rate separately for each age. A life table is necessary to give a good estimate of life expectancy.

Nigeria is one of the most densely populated country in Africa comprising a 923,768 square kilometer area, with a population of 162,471,000 (July 2011 United Nation est.) and its population growth rate 2∙553% that is, the average annual percent change in the population resulting from a surplus or deficit of births over deaths and the balance of migrants entering and leaving a country. The growth rate is a factor in determining how a great burden would be imposed on a country by changing needs of its people for infrastructure, resources and jobs. 

Nigeria’s population growth year-by-year might be best visualized through a comparison of the country’s crude birth and death rates. While the concurrent bump in birth rate and drop in death rate for 1975 through 2000 would be expected to produce swift population growth, the two trends seem to be approaching each other over the long run. As the difference between these rates shrinks, a slowing of population growth should result. However, despite this slowdown, the UN projects a population increase to 210 million individuals in 2025, and 289 million in 2050, the bulk of which will be in the working-age population (UN Population Division 2006). The number of working-age people (aged 15-64) is expected to rise from 75 million in 2005 to 125 million in 2025 and 193 million in 2050. While the number of children (63 million) was not much lower than the number of working-age adults in 2005 (75 million), UN projections suggest that, by 2050, there will be almost triple the number of working-age adults as children, with the ratio of working-age to non-working-age population members just passing by 2050. Nigeria is expected to transition from having 0.90 dependents per worker in 2005 to 0.50 in 2050, as the median age rises from 17.2 to 28 (UN Population Division 2006). As the country’s population ages, however, Nigeria will also face a new challenge of providing for a growing elderly population. 

Gender inequality in Nigeria is among the highest in the world and reflects particular threats to women’s health. The UNDP (2008) ranks the country 140th of 156 countries when comparing the Gender Development Index to the Human Development Index. When comparing male and female life expectancy, Nigeria ranks 177th of 194. High maternal mortality plays a key role in this gap, with unsafe abortions, female genital mutilation (FGM), battery, and sexual assault just some of the factors posing a significant threat to women’s health. While maternal mortality is declining, Nigeria’s rate remains among the highest in the world (WHO 2002). According to UNFPA (2005), maternal mortality fell from 1,000 deaths per 100,000 live births in 1990 to 800 in 2005, but according to the World Bank (2008) the rate was 1100 deaths per 100,000 live births in 2007. In northern Nigeria, women have a 1 in 15 lifetime risk of dying due to a pregnancy-related cause (World Bank/DFID 2005).

At present, statistics is a reliable means of describing accurately the effect of mortality on population growth of any nation. The work of the statistician is no longer confined to gathering and tabulating data, but is chiefly a process of interpreting the information. 

1.1   STATEMENT OF PROBLEM

          Mortality is a serious problem that is of concern not only to a state or a country but also to universe as a whole. It is one of the objectives set by the United Nation to overcome the effect of mortality on population growth for various income categories of countries. Mortality is one of the factors affecting the population of developing countries of the world. Ninety-nine percent of these deaths occur in developing nations. Mortality takes away society’s potential physical, social and human capital.

This research work will look into the problem by analyzing the numerical data on mortality rate in Nigeria with useful recommendations. 



1.2    AIM AND OBJECTIVES OF THE STUDY 

This research work is aimed at studying the effect of mortality on the population growth in Nigeria with the following objectives:

To determine whether the deaths depends on age

To determine whether the deaths depends on gender

To study the relationship between the death rate and age 

To determine whether there is significance difference between male and female deaths

1.3   SIGNIFICANCE OF THE STUDY        

         This research work will be useful in identifying the magnitude and direction of mortality in Nigeria and also assist/help government and non-governmental organizations in planning and implementation of programs.

1.4   SCOPE AND LIMITATION OF THE STUDY

         The period focused for this research work covers a period of ten years from 2002 to 2006 and its limited to the data collected from National Population Commission, NDHS on mortality by age specific death rate and gender.

         The degree of accuracy and precision cannot be ascertained having the above limitation in mind on findings there in, estimate should be expected. This is tolerated since Statistics can also be described as the science of influences of estimates.


1.5   DEFINATION OF TERMS

The terms and concepts to be used are as follows:

NDHS:- National Demographic Household Survey

 Mortality Rate: - is a measure of the number of deaths in a population

cov (x y):-  Covariance between y and x

r:- Correlation coefficient between y and x

Ho: There is no significant difference

H1: There is significant difference

:- Level of significance  

 

CHAPTER TWO

2.0    LITERATURE REVIEW

According to Jamison (2003), improvement of life expectancy especially in developing countries like Nigeria and the sharp drop in mortality can be attributed to the huge investment in the health sector by most government especially in the developing countries.

According to Fapohunda (1979), the size of any population is correlated with the level of fertility and mortality. It is positively correlated in the first case and negatively correlated with the later. She argues that the Nigerian population is growing at a high rate of 2.5% per year due to a high fertility rate and a decreasing mortality rate. Fapohunda further argued that since the population is growing rapidly due to a high fertility rate and a declining mortality rate, an increasing proportion of the population is under 15years of age and is for the most economically inactive. They are basically dependent on the economically active group.

According to Ann (2008), the impact of the disease on the economy and economic development efforts as seen in reduced productivity caused by workers illness and its financial burden on individuals and government cannot be over emphasized. Due to its widespread nature, precise estimates of the mortality and morbidity of malaria are often hard to come by. Unsurprisingly, there are no accurate statistical estimates of deaths caused by malaria in Nigeria.

Jamison (2003), in his study in the 2003 world development report, attributed several causes of death to environmental habit and social factors. According to him, many cities suffer from air pollution caused by industries, power plants, road transport and domestic use of coal. He stressed that atmospheric pollution is a major cause of respiratory disease in urban Centre. Jamison further stressed that inadequate environmental sanitation and impure water supply are major causes of tuberculosis and typhoid diseases. He claims that about1.3 billion people in developing countries lack access to clean and portable water. Other causes of deaths according to medical scientists are those that arise from habits such as tobacco, smoking, alcohol and illegal drugs use or drug abuse. Although tobacco is in legal use everywhere in the world, but it causes far more deaths than all other “psycho-active” substances.

Law (1998) attempted to identify and quantify the factors responsible for the differences in mortality between affluent and deprived areas, the north and the south, and urban and rural areas, in an analysis of cause specific mortality in the 403 local authority districts in England and Wales. Reported that: The excess mortality from other diseases closely matched that predicted from differences according to deprivation and latitude in smoking, heavy alcohol consumption, Helicobacter pylori infection, and temperature, and thus could be attributed to these causes. About 85% of the overall excess mortality with deprivation was attributable to heavier smoking and 6% to heavier alcohol consumption, but diet varied little. Deaths more directly related to deprivation accounted for an estimated 12% of the excess deaths, but variation in provision and uptake of healthcare services only 1%. The more direct material effects of deprivation contribute to the variation in mortality but is particularly important with respect to differences in morbidity. Of the difference in mortality with latitude, about 45% was attributable to differences in smoking, and 25% to climate (mainly the association of cardiovascular and respiratory disease with cold). The differences with urbanization were mainly because of smoking. 

Jessop (1996) computed SMRs for all causes of death, for bronchitis and asthma (ICD9 codes 490-493), and for accident, violence, and poisoning (ICD9 codes 800-999), for members of a longitudinal study representing a quasi-random 1% sample of the population of England and Wales. Deprivation was classified by an electoral ward deprivation score and by home and car ownership. It was found that:  There was an association between deprivation and mortality that was clear for all-cause mortality, more noticeable for respiratory disease, and less clear for deaths from accident, violence, and poison. In general, the results showed a remarkable similarity between metropolitan and non-metropolitan areas. This study does not support the hypothesis that the relationship between mortality and deprivation differs between residents of metropolitan and non-metropolitan areas of England and Wales. 

Morris (1996) carried out an ecological study relating measures of mortality to local rates of educational attainment at age 15/16 years and scores on the Department of the Environment's index of local conditions. The main findings were: Educational attainment was closely associated with all cause, coronary, and infant mortality and strongly associated with the index of local conditions. This social index was also closely associated with all the measures of mortality. In multiple regression, the social index was the stronger correlate of all cause mortality but for coronary and infant mortality, educational attainment remained highly statistically significant. Area levels of both educational attainment and deprivation-affluence are strong correlates of local mortality rates in England. In these analyses educational attainment may be indexing the general cultural level of a community. 

Bentham (1995) assessed the level of similarity between geographical variation in limiting long-term illness and the distribution of mortality rates and of deprivation. The main findings were: The geographical pattern of limiting long-term illness shows many similarities with that of mortality but there are also some differences. Both are positively associated with indicators of social deprivation, with limiting long-term illness tending to show stronger correlations, particularly in the elderly. Most of Wales and many industrial areas of northern England have higher rates of long term illness than would be expected from their mortality rates, while much of south eastern England has lower than expected rates. However, further assessment of the reliability of these data on self-reported morbidity is required. 

Carstairs (1995) reported that: The link between deprivation and health has been clearly demonstrated in a number of studies, with populations living in deprived areas exhibiting levels of mortality, particularly below the age of 65, which vastly exceed those in affluent areas. In the decade 1981-91, these differentials increased in Scotland and the Northern Health Region and inequalities in health widened. Analysis shows that particular causes of death and sites of cancer are more likely to reflect the influence of socio-economic factors. The work so far mostly shows the associations between these factors and health measures and more investigation is required into the determinants of health. The authors conclude that area measures of deprivation can prove valuable in examining differentials in health and death. 

Sloggett (1998) reported that: Without adjusting for personal circumstances, the risk of premature death (before age 70) and the risk of long term limiting illness in 1991, but not risk of stillbirth, showed a clear, significant, and approximately linear association with social deprivation of ward of residence in 1981.When adjustment is made for personal disadvantage the associations with local area deprivation were all attenuated, especially for those living in the more deprived areas. Residence in more deprived areas was more strongly associated with long- term illness than with mortality. These associations seem to be largely because residence in more deprived areas is associated with personal disadvantage.

Davey-Smith (1998) reported that:  Over 21 years of follow up, 1,639 of the men died.  Mortality from all causes and from three broad causes of death groups (cardiovascular disease, malignant disease, and other causes) showed similar associations with social class and education. For all cause of death groups, men in manual social classes and men who terminated full time education at an early age had higher death rates. Cardiovascular disease was the cause of death group most strongly associated with education, while the non-cardiovascular non-cancer category was the cause of death group most strongly associated with adulthood social class.  The graded association between social class and all-cause mortality remains strong and significant within education strata, whereas within social class strata the relation between education and mortality is less clear. As a single indicator of socio-economic position occupational social class in adulthood is a better discriminator of socio-economic differentials in mortality than is education. This argues against interpretations that see cultural, rather than material, resources as being the key determinants of socio-economic differentials in health. The stronger association of education with death from cardiovascular causes than with other causes of death may reflect the function of education as an index of socio-economic circumstances in early life, which appear to have a particular influence on the risk of cardiovascular disease. 

Dolk (1995) calculated standardized morbidity ratios for cancers in 1981 and standardized mortality ratios for all-cause mortality between 1982 and 1985. A deprivation index was used to measure gradients of deprivation according to the distance from industrial sites. Findings include: Strong relationships were found between all-cause mortality and deprivation quintile. The relationship between deprivation, urban/rural status, and mortality is complex and is confounded by region. Mortality tends to be higher in urban than in rural areas within quintiles of deprivation. The main problems in the interpretation of the deprivation index may be its limited correlation with the risk factors of interest and its concentration on present rather than past socio-economic status. There is potential for important socio-economic confounding in small area studies of environmental pollution and health where the health outcome under examination has a strong relationship to socio-economic status and where the putative excess risk due to pollution may be small. One method of controlling for confounding is to use an ecological measurement of deprivation in small areas, and to adjust for deprivation by indirect standardization. However, residual socio-economic confounding can be expected, which may seriously complicate the interpretation of small area studies. 

Gognalons (1999) reported that: There were several social prognostic factors of mortality independent of health and occupational status with relative risks greater than 3.0 including:  period of unemployment during life time, feeling of not demonstrating initiative in the occupational setting, non-participation in social activities. The results suggest that differential mortality determined by occupational status can be explained in part by factors that are characteristic of "life style", social dynamics, occupational context, and ruptures during the course of occupational life. 

Findings from Hunt (1991) include: The proportion of deaths that occurred at home decreased from 55.6% in 1910 to 26.2% in 1970 and thereafter about a quarter of all deaths occurred at home. By 1970 over two-thirds of all deaths occurred in hospitals; after 1970 death has been transferred from hospitals to nursing homes and inpatient hospices. Mortality patterns are determined by social and demographic characteristics availability of hospital and nursing home beds changes to health insurance schemes emergence of hospice care and related services. 

Huff (1999) adopted a small area ecological study design and calculated ward level (n=591) standardized ratios (population aged <75) for specific causes of death and limiting long-term illness. Deprivation was measured using the Townsend Index. It was found that: Wide variations in mortality and illness were evident at ward level, being widest for accident mortality (standardized mortality range 0-508).  Stroke mortality accounted for the largest proportion of wards with standardized mortality ratios over 125 (36.2%). Relative deprivation correlated strongly with limiting long-term illness (r=0.82) and all-cause mortality (r=0.68) across Trent, and in both urban and rural environments. Wide health inequalities were evident in Trent and the association between deprivation and health was of a similar magnitude in urban and rural wards. This small area approach allows health authorities access to ward level information in order to inform key debate on tackling health inequalities and distributing resources in relation to need. 

Johnson (2000) reported that: For persons aged 45-64, each of the non-married groups generally showed statistically significant increased risk compared to their married counterparts (RR for white males, 1.24-1.39; white females, 1.46-1.49; black males, 1.27-1.57; and black females, 1.10-1.36). Older age groups tended to have smaller RRs than their younger counterparts. Elevated risk for non-married females was comparable to that of non-married males. For cardiovascular disease mortality, widowed and never-married white males ages 45-64 showed statistically significant increased RRs of 1.25 and 1.32, respectively, whereas each non-married group of white females showed statistically significant increased RRs from 1.50 to 1.60. RRs for causes other than cardiovascular diseases or cancers were high (for white males ages 45-64: widowed 1.85; divorced/separated, 2.15; and never-married, 1.48).  Labor force status was important in determining the elevated risk of non-married males compared to non-married females by race. 

Birch (1996) assessed the levels of correlation of indicators based on mortality data, socio-economic data, and combined data with a standardized indicator of self-assessed health, in a cross-sectional analysis of population data for the 15 health regions in Quebec. The following results were reported:  Variations in scores of a proxy based on socio-economic data among regions explain 37% of the observed variation in self-assessed health, 4% more than the level of variation explained by the standardized mortality rate scores. A weighted combination of both mortality and socio-economic based proxies explains 56% of variation in self-assessed health. Justification of "deprivation weights" reflecting variations in socio-economic status among populations should be based on empirical support concerning the performance of such weights as proxies for relative levels of need among populations. The socio-economic proxy developed in this study provides a closer correlation to the self-assessed health of the populations under study than the mortality-based proxy. The superior performance of the combined indicator suggests that the development of social deprivation indicators should be viewed as a complement to, as opposed to a substitute for, mortality-based measures in needs-based resource allocation exercises. 

Drever (1995) used a modified form of the Department of Environment's 1991 deprivation index to study interaction between socio-economic and geographic variables in mortality in more than 350 English local authorities in 1989-93. The main findings reported were: The familiar geographic pattern of higher mortality in the north and west and lower mortality to the south and east of the country has continued into the 1990s and there has been no significant widening or narrowing in the mortality gap between the worst and the best regions during the 1980s. The local authorities with higher mortality are still predominantly in urban areas. There is a very strong relationship between mortality and deprivation at the local authority level measured by the new index, with a tendency for higher mortality to be associated with greater deprivation. This relationship is most marked for males, but is still strong for females. 




 

CHAPTER THREE

3.0 RESEARCH METHODOLOGY

Research methodology is a process to carry out a research. This means the method used for data collection, source of data, type of data and techniques used for data analysis. There are different approaches to the method of collecting data about which a research is base; this depends on the type of research to be carried out and the type of data needed for the subject matter.

3.1 DATA COLLECTION METHODS

There are many methods to collect data, depending on the research design and the methodologies employed. Some of the common methods are questionnaires, interview and observation.

3.2 TYPES OF DATA

Data can be divided into two types, namely quantitative and qualitative. Quantitative data is numerical in nature and can be mathematically computed. Quantitative data measure uses different scales, which can be classified as nominal scale, ordinal scale, interval scale and ratio scale. Qualitative data are mostly non-numerical and usually descriptive or nominal in nature. This means the data collected are in the form of words and sentences.

3.3 SOURCES OF DATA

Generally we can collect data from two sources, primary sources and secondary sources. Primary data are also known as raw data. Data are collected from the original source in a controlled or an uncontrolled environment. Example of a controlled environment is experimental research where certain variables are being controlled by the researcher. On the other hand, data collected through observation or questionnaire survey in a natural setting are examples data obtained in an uncontrolled environment. Secondary data are data obtained from secondary sources such as records, reports, books, journals, documents, magazines, the web and more.

The data used for this research work is secondary data. The data was collected from National Population Commission, NDHS on mortality by age specific death rate and gender. The data covers the period: 2002-2006

3.4 METHOD OF ANALYSIS

The statistical techniques used to analyze our data are Chi-square test of independence, Correlation Coefficient and T-test

3.4.1 CONTINGENCY TABLE

A contingency table is a display for two categorical variables. Its rows list the categories of one variable and its columns list the categories of the other variable. Each entry in the table is the frequency of the cases in the sample with certain outcome on the two variables. Each row and column combination in a contingency table is called a cell. The process of taking data and finding the frequency for each cell of a contingency table is referred to as cross-tabulation of the data. For a contingency table that has r rows and c columns, you can generalize the χ2 test as a test of independence for two categorical variables.

3.4.2 THE CHI-SQUARED TEST OF INDEPENDENCE

The test statistic for the test of independence summarizes how close the observed cell counts fall to expected cell counts. Symbolized by, it is called the Chi-squared statistic introduced by British statistician Karl Pearson in 1900, it is sometime called “Pearson’s chi-square statistic”. Pearson’s chi-square statistic is used to test independence between the row and column variables. Independence means that knowing the value of the row variable does not change the probabilities of the column variable (and vice versa). The chi-square statistic summarizes how far the observed cell counts in a contingency table fall from expected cell counts for a null hypothesis. Its formula is

    

When H0: independence is true, the observed and expected counts tend to be close for each cell. Thenhas a relative small value. If H0: is false, at least some observed and the expected counts tend to be far apart. For a cell where this happens its value of  tend to be large. Then  has a relative large value.

The chi-square test of independence tests the hypothesis that the variables are independent of each other. The computed value is compared with the tabulated value at desired level of significance. Alternatively, the p-value is compare with  and reject H0 if 


PROPERTIES OF CHI-SQUARE

It falls on the positive part of a real number line. The chi-square test statistic cannot be negative, since it sums squared difference divided by positive expected frequencies. The minimum value of chi-square test statistic is equal to zero, and would occur if observed count is equal to expected count in each cell.

 The precise shape of the distribution depends on the degree of freedom.  For testing independence in a table with r rows and c columns (called r by c table), the formula for the degree of freedom is df = (r-1) (c-1).

 It is skewed to the right. As degree of freedom increase, the skew lessens and the chi-square curve becomes more bell-shaped and spread out.

The larger the value of chi-square test statistic, the greater the evidence against the H0: independence thus, the p-value equals the right tail probability. It measures the probability that the chi-square test statistic for a random sample would be larger than the observed value, if the variables are truly independent. (Agresti and Franklin 2007)

3.4.4 FIVE STEPS FOR CHI-SQUARE TEST OF INDEPENDENCE

Assumptions: the expected count greater than or equal to 5 in all cell.

Hypotheses: H0: The two variables are independent, Ha: The two variables are dependent (associated).

P-value: Right –tail probability above observed chi-square value, for the chi-square distribution with df = (r-1) (c-1).

Conclusion: Report p-value and interpret in context, if a decision is needed, reject H0 if p-value is less than or equal to significance level. (Agresti and Franklin 2007).

3.4.5 TESTS OF HYPOTHESES FOR CHI-SQUARE TEST OF INDEPENDENCE

The hypothesis formulated for this test has to be tested by using information from analyzed data as basis to determine its validity.

Ho: The two variables are independent (that is, there is no relationship between them)

H1: The two variables are dependent (that is, there is relationship between them

3.4.6 CORRELATION COEFFICIENT

Correlation coefficient measures the direction and strength of the linear relationship between the two variables, if the linear relationship between Y and X is positive (as X increases Y also increases) or conversely, if the relationship between Y and X is negative (as X increases Y decreases).

  

Thus, Cor(Y, X) can be interpreted either as the covariance between the standardized variables or the ratio of the covariance to the standard deviations of the two variables. The correlation coefficient is symmetric, that is, Cor(Y, X) = Cor(X, Y). The Cor(Y, X) a useful quantity for measuring both the direction and the strength of the relationship between Y and X. The magnitude of Cor(Y, X) measures the strength of the linear relationship between Y and X. The closer Cor(Y, X) is to 1 or -1, the stronger is the relationship between Y and X. The sign of Cor (Y, X) indicates the direction of the relationship between Y and X. That is, Cor (Y, X) > 0 implies that Y and X are positively related. Conversely, Cor (Y, X) < 0, implies that Y and X are negatively related. Note, however, that Cor (Y, X) = 0 does not necessarily mean that Y and X are not related. It only implies that they are not linearly related because the correlation coefficient measures only linear relationships. In other words, the Cor (Y, X) can still be zero when Y and X are nonlinearly related. The test of hypothesis about the correlation coefficient, the hypothesis Ho: means that there is no linear relationship between the variables. (Samprit and Hadi 2006)

3.4.7 TESTS OF HYPOTHESES FOR CORRELATION COEFFICIENT

The hypothesis formulated for this test has to be tested by using information from analyzed data as basis to determine its validity.

Ho: There is no relationship between the two variables

H1: There is relationship between the two variables

3.4.8 THE T-TEST

The t-test is distributed as a Student’s t with (n - 2) degrees of freedom. The test is carried out by comparing the observed value with the appropriate critical value obtained from the t-table, which is  where specified significance level. Note that we divide  by 2 because we have a two-sided alternative hypothesis. Accordingly, H0: is to be rejected at the significance level if


Where  denotes the absolute value of t1. A criterion equivalent to that in is to compare the p-value for the t-test with  and reject H0 if


Where  called the p-value, is the probability that a random variable having a Student t distribution with (n - 2) is greater than  (the absolute value of the observed value of the t-test). The p-value is the sum of the two shaded areas under the curve. The p-value is usually computed by statistical packages. (Samprit and Hadi 2006)

 

3.4.9 TESTS OF HYPOTHESES FOR T-TEST

The hypothesis formulated for this test has to be tested by using information from analyzed data as basis to determine its validity.

Ho: The means of genders are equal

H1: The means of genders are not equal

 

CHAPTER FOUR

4.0 DATA PRESENTATION AND ANALYSIS

4.1 DATA PRESENTATION

The data presented below are data of mortality obtained from National Population Commission

Age Group

      2002

      2003

      2004

      2005

      2006

Grand

Total



M

F

M

F

M

F

M

F

M

F



Less than 1

70

44

153

63

128

41

16

7

227

67

1976


01-04

344

253

371

200

174

148

121

58

137

107

10798


05-09

131

94

75

57

89

79

414

171

48

53

4669


10-14

96

61

85

41

58

36

23

16

63

35

2568


15-19

146

110

77

66

57

59

32

22

64

55

3242


20-24

258

246

206

222

142

112

78

71

129

129

5928


25-29

463

453

371

423

339

261

143

153

268

268

9996


30-34

604

424

443

412

453

312

230

189

381

325

10850


35-39

891

427

601

387

551

328

280

203

615

363

12067


40-44

870

335

728

326

609

288

379

169

792

340

11728


45-49

832

281

733

267

795

270

487

177

917

304

10998


50-54

804

291

750

224

814

276

457

154

899

245

11261


55-59

665

222

646

186

717

215

474

122

872

236

8911


60-64

726

242

670

191

712

217

446

93

922

228

10142


65-69

624

207

482

162

554

185

380

116

762

196

8298


70-74

649

233

485

147

555

178

399

115

741

181

8472


75-79

382

165

310

137

392

134

254

74

546

119

5715


80-84

344

172

258

116

318

126

190

69

431

133

5276


85+

520

326

528

298

549

270

322

146

565

234

10675


Total

9419

4586

7972

3925

8006

3535

5225

2125

9379

3618

153570


Source: National Population Commission, ND

 

4.2 DATA ANALYSIS 

Table 4.1 Contingency Table for Age Category and Year


Year

Total



2002

2003

2004

2005

2006



Age Category

less than 1

114

216

169

23

294

816



01-04

597

571

322

179

244

1913



05-09

225

132

168

585

101

1211



10-14

157

126

94

39

98

514



15-19

256

143

116

54

119

688



20-24

504

428

254

149

258

1593



25-29

916

794

600

296

536

3142



30-34

1028

855

765

419

706

3773



35-39

1318

988

879

583

978

4746



40-44

1205

1054

897

548

1132

4836



45-49

1113

1000

1065

664

1221

5063



50-54

1095

974

1090

611

1144

4914



55-59

877

832

932

596

1108

4345



60-64

968

861

929

539

1150

4447



65-69

831

644

739

496

958

3668



70-74

882

632

733

514

922

3683



75-79

547

447

526

328

665

2513



80-85

516

374

444

259

564

2157



85+

846

826

819

468

799

3758


Total

13995

11897

11541

7350

12997

57780



The above table shows the death rate of specific age group where rows represent the age category against the columns which represent years. Each entry in the table is the death rate of specific age group against the year. These are the observed values which will lead us to calculate the expected values. Each row and column combination in a contingency table is called a cell. The process of finding the expected value for each cell of a contingency table is referred to as cross-tabulation of the data. 



 

Table 4.2 Expected Count and Observed Count for Age Category and Year


2002

2003

2004

2005

2006



Age Category

less than 1

Count

114

216

169

23

294

816




Expected Count

197.6

168.0

163.0

103.8

183.6

816.0



1-4

Count

597

571

322

179

244

1913




Expected Count

463.4

393.9

382.1

243.3

430.3

1913.0



5-9

Count

225

132

168

585

101

1211




Expected Count

293.3

249.3

241.9

154.0

272.4

1211.0



10-14

Count

157

126

94

39

98

514




Expected Count

124.5

105.8

102.7

65.4

115.6

514.0



15-19

Count

256

143

116

54

119

688




Expected Count

166.6

141.7

137.4

87.5

154.8

688.0



20-24

Count

504

428

254

149

258

1593




Expected Count

385.8

328.0

318.2

202.6

358.3

1593.0



25-29

Count

916

794

600

296

536

3142




Expected Count

761.0

646.9

627.6

399.7

706.8

3142.0



30-34

Count

1028

855

765

419

706

3773




Expected Count

913.9

776.9

753.6

480.0

848.7

3773.0



35-39

Count

1318

988

879

583

978

4746




Expected Count

1149.5

977.2

948.0

603.7

1067.6

4746.0



40-44

Count

1205

1054

897

548

1132

4836




Expected Count

1171.3

995.7

965.9

615.2

1087.8

4836.0



45-49

Count

1113

1000

1065

664

1221

5063




Expected Count

1226.3

1042.5

1011.3

644.0

1138.9

5063.0



50-54

Count

1095

974

1090

611

1144

4914




Expected Count

1190.2

1011.8

981.5

625.1

1105.4

4914.0



55-59

Count

877

832

932

596

1108

4345




Expected Count

1052.4

894.6

867.9

552.7

977.4

4345.0



60-64

Count

968

861

929

539

1150

4447




Expected Count

1077.1

915.6

888.2

565.7

1000.3

4447.0



65-69

Count

831

644

739

496

958

3668




Expected Count

888.4

755.2

732.6

466.6

825.1

3668.0



70-74

Count

882

632

733

514

922

3683




Expected Count

892.1

758.3

735.6

468.5

828.5

3683.0



75-79

Count

547

447

526

328

665

2513




Expected Count

608.7

517.4

501.9

319.7

565.3

2513.0



80-84

Count

516

374

444

259

564

2157




Expected Count

522.5

444.1

430.8

274.4

485.2

2157.0



85+

Count

846

826

819

468

799

3758




Expected Count

910.2

773.8

750.6

478.0

845.3

3758.0


Total

Count

13995

11897

11541

7350

12997

57780



Expected Count

13995.0

11897.0

11541.0

7350.0

12997.0

57780.0



The above table shows the expected and observed death rate of each specific age group against the years. The entries in the table are the expected and observed death rate of specific age group against the year. These expected values in each cell is as a result of the ratio of the row total by column total to the grand total which will lead us to calculate the χ2 test as a test of independence for the two variables. Each row and column combination in a contingency table is called a cell. The process of finding the expected value for each cell of a contingency table is referred to as cross-tabulation of the data. For a contingency table that has rows and columns, you can generalize the χ2 test as a test of independence for the two variables.


Table 4.3 Chi-square Test of Independence for Age Category and Year 




Value

df

Asymp. Sig. (2-sided)


Pearson Chi-Square

2576.650a

72

.000


Likelihood Ratio

2149.379

72

.000


Linear-by-Linear Association

273.513

1

.000


N of Valid Cases

57780




a. 0 cells (0.0%) have expected count less than 5. The minimum expected count is 65.38.



The above table shows the χ2 test of independence for the two variables which is the sum of squares of the difference between the observed values and the expected values to the observed values. Since, the p-value of 0.00 is less than 0.01 level of significance, therefore the null hypothesis is rejected, and this implies that there is linear relationship between deaths and age category.




 

Table 4.4 Contingency Table for Age Category and Gender


Gender

Total



Male

Female



Age Category

less than 1

594

222

816



01-04

1147

766

1913



05-09

757

454

1211



10-14

325

189

514



15-19

376

312

688



20-24

813

780

1593



25-29

1584

1558

3142



30-34

2111

1662

3773



35-39

3038

1708

4746



40-44

3378

1458

4836



45-49

3764

1299

5063



50-54

3724

1190

4914



55-59

3364

981

4345



60-64

3476

971

4447



65-69

2802

866

3668



70-74

2829

854

3683



75-79

1884

629

2513



80-85

1541

616

2157



85+

2484

1274

3758


Total

39991

17789

57780



The above table shows the death rate of specific age group where rows represent the age category against the columns which represent gender. Each entry in the table is the death rate of specific age group against the gender. These are the observed values which will lead us to calculate the expected values. Each row and column combination in a contingency table is called a cell. The process of finding the expected value for each cell of a contingency table is referred to as cross-tabulation of the data. 




Table 4.5 Expected Count and Observed Count for Age Category and Gender


Male

Female



Age Category

less than 1

Count

594

222

816




Expected Count

564.8

251.2

816.0



1-4

Count

1147

766

1913




Expected Count

1324.0

589.0

1913.0



5-9

Count

757

454

1211




Expected Count

838.2

372.8

1211.0



10-14

Count

325

189

514




Expected Count

355.8

158.2

514.0



15-19

Count

376

312

688




Expected Count

476.2

211.8

688.0



20-24

Count

813

780

1593




Expected Count

1102.6

490.4

1593.0



25-29

Count

1584

1558

3142




Expected Count

2174.7

967.3

3142.0



30-34

Count

2111

1662

3773




Expected Count

2611.4

1161.6

3773.0



35-39

Count

3038

1708

4746




Expected Count

3284.8

1461.2

4746.0



40-44

Count

3378

1458

4836




Expected Count

3347.1

1488.9

4836.0



45-49

Count

3764

1299

5063




Expected Count

3504.2

1558.8

5063.0



50-54

Count

3724

1190

4914




Expected Count

3401.1

1512.9

4914.0



55-59

Count

3364

981

4345




Expected Count

3007.3

1337.7

4345.0



60-64

Count

3476

971

4447




Expected Count

3077.9

1369.1

4447.0



65-69

Count

2802

866

3668




Expected Count

2538.7

1129.3

3668.0



70-74

Count

2829

854

3683




Expected Count

2549.1

1133.9

3683.0



75-79

Count

1884

629

2513




Expected Count

1739.3

773.7

2513.0



80-84

Count

1541

616

2157




Expected Count

1492.9

664.1

2157.0



85+

Count

2484

1274

3758




Expected Count

2601.0

1157.0

3758.0


Total

Count

39991

17789

57780



Expected Count

39991.0

17789.0

57780.0


The above table shows the expected and observed death rate of each specific age group against the gender. The entries in the table are the expected and observed death rate of specific age group against the gender. These expected values in each cell is as a result of the ratio of the row total by column total to the grand total which will lead us to calculate the χ2 test as a test of independence for the two variables. Each row and column combination in a contingency table is called a cell. The process of finding the expected value for each cell of a contingency table is referred to as cross-tabulation of the data. For a contingency table that has rows and columns, you can generalize the χ2 test as a test of independence for the two variables.


Table 4.6 Chi-square Test of Independence for Age Category and Gender




Value

df

Asymp. Sig. (2-sided)


Pearson Chi-Square

2041.657a

18

.000


Likelihood Ratio

1990.978

18

.000


Linear-by-Linear Association

746.402

1

.000


N of Valid Cases

57780




a. 0 cells (0.0%) have expected count less than 5. The minimum expected count is 158.25.



The above table shows the χ2 test of independence for the two variables is the sum of squares of the difference between the observed values and the expected values to the observed values.  Since, the p-value of 0.00 is less than 0.01 level of significance, therefore the null hypothesis is rejected, and this implies that there is linear relationship between deaths and gender. 


 

Table 4.7 Pearson Correlations between Death and Gender



Death

Gender


Death

Pearson Correlation

1

-.504**



Sig. (2-tailed)


.000



N

190

190


Gender

Pearson Correlation

-.504**

1



Sig. (2-tailed)

.000




N

190

190




The above table shows the correlation between death rate and gender. Since, the p-value of 0.00 is less than 0.01 level of significance, therefore the null hypothesis is rejected, and this implies that there is linear relationship between deaths and gender.




Table 4.8 Pearson Correlations between Death and Age Group


Death

Age Group


Death

Pearson Correlation

1

.369**



Sig. (2-tailed)


.000



N

190

190


Age Group

Pearson Correlation

.369**

1



Sig. (2-tailed)

.000




N

190

190



The above table shows the correlation between death rate and age category. Since, the p-value of 0.00 is less than 0.01 level of significance, therefore the null hypothesis is rejected, and this implies that there is linear relationship between deaths and age group


 

Table 4.9 Descriptive Statistics of Gender


Gender

N

Mean

Std. Deviation

Std. Error Mean


Death

Male

95

420.96

263.023

26.986



Female

95

187.25

108.407

11.122




Table 4.10 Test Significance between the Genders


Levene's 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


Death

74.382

.000

8.007

188

.000

233.705

29.188

176.128

291.283



The above table shows the test of significance between the genders, its test whether there is significance difference between the males and females death rate. Since, the significance value (p-value) of the test statistic is 0.00 is less than 0.01 level of significance, we reject the hypothesis that the means of the genders are equal.

 

CHAPTER FIVE

5.0 SUMMARY, CONCLUSION AND RECOMMENDATION

5.1 SUMMARY 

            This research work tends to study the effect of mortality rate in Nigeria. Mortality rate is the number of deaths in a given area or population. Nigeria is one of the most densely populated country in Africa with a population of 162,471,000 (July 2011 United Nation est.). The methods used for the analysis were Chi-square test of independence, Correlation Coefficient and T-test.

             From Table 4.3, the Chi-square test of independence for the two variables (age category and year) which is the sum of squares of the difference between the observed values and the expected values to the observed values. Since, the p-value of 0.00 is less than 0.01 level of significance, therefore the null hypothesis is rejected, and this implies the two variables are dependent that is, there is linear relationship between deaths and age category. Also, Table 4.6 shows the Chi-square test of independence for the two variables deaths and gender.  Since, the p-value of 0.00 is less than 0.01 level of significance, therefore the null hypothesis is rejected, and this implies that there is linear relationship between deaths and gender.

From Table 4.7 shows the correlation between death rate and gender. Since, the p-value of 0.00 is less than 0.01 level of significance, therefore the null hypothesis is rejected, and this implies that there is linear relationship between deaths and gender. Also, Table 4.8 shows the correlation between death rate and age category. Since, the p-value of 0.00 is less than 0.01 level of significance, therefore the null hypothesis is rejected, and this implies that there is linear relationship between deaths and age group.


From Table 4.10, shows the test of significance between the genders, its test whether there is significance difference between the males and females death rate. Since, the significance value (p-value) of the test statistic is 0.00 is less than 0.01 level of significance, we reject the hypothesis that the means of the genders are equal.


5.2 CONCLUSION

         Based on the findings, the differences between the deaths of the age groups are due to chance variation which implies that there is significance difference between the deaths of age groups and gender. 

5.3 RECOMMENDATION

Human capital is one of the most essential features of any developed country. It is paramount important for government to have a full coverage of its mortality rate for policy implementation, therefore the government should organize workshops to enlighten the general public on personal hygiene and improve the health care system. 

 

REFERENCES

Agesti, A. (1992). A Survey of Exact Inference for Contingency Table. John Wiley & Sons Inc., Hoboken New Jersey, 131-141p.

Agresti, A. (1996). An Introduction to Categorical Data Analysis. A John Wiley & Sons Inc., New Jersey, 3, 27p.

Agresti, A. and Franklin, C. (2007). The Art and Science of Learning from Data. Upper Saddle River, New Jersey. 88, 484p 

Alvin, C. R. and Bruce, S. G. (2008).  Linear Models in Statistics. York, Inc., 1-16.  Business media, New York, 1-27.                               

Ann, F. (2008). Population, Resources and Environment (http://www.Voxfux.com/features/malthusian_theory.htm).

Cyprian, A. O. (2009).  An Introduction to Applied Statistical Methods. (8nd ed), Modern printers, Emene, Enugu, 374-390. 

John, O. R., Sastry, G. P., and David, A. D. (1998).  Applied Regression Analysis: A research tool. (2nd ed), Springer Verlag New

Samprit, C., and Ali S. H. (2006). Regression by Example. (4nd ed), John Wiley & Sons Inc., Hoboken New Jersey, 21-39.

Sanford, W. (2005).  Applied Linear Regression. (3rd ed), A John Wiley & Sons Inc., Hoboken New Jersey, 21-31.

Simon, J. S. (2009).  A Modern Approach to Regression. Springer (2nd ed), A John Wiley &   Sons Inc., Hoboken New Jersey, 1-3.


0/Post a Comment/Comments