CHAPTER ONE
INTRODUCTION
BACKGROUND OF STUDY
Malaysia was well known as having the highest obesity prevalence in Southeast Asia. This was proved in Economist Intelligence Unit’s “Tackling Obesity in Asean” report, which covered Malaysia, Singapore, Indonesia, Thailand, the Philippines and Vietnam. In Malaysia the rates of obesity among adolescents has been significantly rise at a dramatic rate along with the prevalence of weight-related diseases.
The adolescent obesity can cause intermediate risk factor of non-communicable diseases such as diabetes mellitus, cardiovascular disease and hypertension. Moreover, adolescence obesity also can cause physiological disorder and premature death (Rofey et al., 2009). Therefore, it is very important to determining the factors that cause obesity in childhood, a period of rapid growth and development. There are lots of factors affect the occurrence of obesity (Faith, Scanlon, Birch, Francis ; Sherry, 2004).
Firstly, eating behaviors may be defined as biological and behavioral processes directed towards meeting requirements for health and growth, and they evolve during the first years of life (Savage, Fisher, ; Birch, 2007). Food responsiveness (FR), emotional overeating (EOE), enjoyment of food (EOF) and food fussiness (FF) are among several types of eating behavior (Wardle, Guthrie, Sanderson, ; Rapoport, 2001).
Moreover, there is study that conclude there is positive correlation for adolescent obesity and parent income in very low-income families. But, Jo, 2014 indicate the negative association between parent income and BMI is especially significant among high-BMI children and the difference in obesity rates between children from low- and high-income families increases as children age. Body mass index (BMI) of more than 25 means the person is overweight (WHO Expert Consultation, 2004). BMI can be defined as weight in kilograms divided by height in
meters squared.Next, over recent years, there has been an increasing evidence that proof that disorders of sleep might be an associated with the improvement of weight in adolescent. In spite of the fact that it is perceived that there is no ‘magic number’ for the ideal hours of rest, the national rest establishment prescribes that nominal rest is between 7 and 9 hour per day. However, according to (Theorell-Haglöw, Berglund, Janson ; Lindberg, 2012), there may be differences in factors that influence to obesity in short sleepers versus long sleepers.
Besides, exercise is one of the factor that contributing to high rate of adolescent obesity. Family, social environment, and peers, affect adolescents’ exercise behaviors. Interventions including healthy diets and adequate exercise are effective for controlling weight during adolescence (Berkowitz et al., 2013; Epstein, Paluch, Roemmich, & Beecher, 2007; Savoye et al., 2011).
Furthermore, according to Health and Social Care Information Centre, 2012/13 indicate that National Child Measurement Programme (NCMP) found that children from urban area has higher level of obesity than rural area. Along with rapid economic development and urbanization, there have been great changes in lifestyle and diet, especially in rural area that can prevalence obesity (Zhou, Wang, Zhang, Zhang, & Wang, 2016).
Notably, most of studied assessing parenting feeding style that affect child weight have been conducted in Malaysia, but none of the studies were found which conducted a comprehensive assessment of eating behavior, parent income, disorder of sleep and exercise that affect obesity among adolescent in Kelantan. The purpose of this study was to determine how these factors affect obesity among adolescent between rural and urban school for Form Four and Form Five in Kota Bharu, Kelantan.
PROBLEM STATEMENT
Obesity can be defined as “abnormal or excessive fat accumulation that presents a risk to health” by the World Health Organization (WHO), is one of the most significant health problems in developed and developing countries (World Health Organization, 2000). Next, the previous researchers also found that majority of the children nowadays are obese or overweight around the world. The frequency of exercise, duration of sleep and also family income could give a great influence towards the status of obesity among adolescent. The previous authors also found that emotional eating appears to be associated with overeating and eating foods high in energy density among adolescents (Nguyen-Michel, Unger, & Spruijt-Metz, 2007). In addition, obesity problem was also prevalent among Malaysian adolescent from rural and urban secondary school that has been associated with certain behavioural factors (Fadzlina et al., 2014). Furthermore, another study that has been conducted also shows that 30% of overweight adolescent came from Kota Bharu district (Sakinah, Ting, Rosniza & Jayah, 2012). These articles, depicted that obesity among the adolescent is a serious problem in Malaysia. Thus, identifying the types of health behaviors among adolescents is important within this critical age, because these behaviors can translate to habits which continue into adulthood. Obese youth are more likely to have risk factors for cardiovascular disease, such as high cholesterol or high blood pressure. It is important to determine the risk factors that affect obesity in childhood, in order to rise generations of healthy children.
RESEARCH OBJECTIVES
This study focus on the following objectives:
To identify the percentage of adolescent who were obese and non-obese between rural and urban secondary school.
To study the relationship between the risk factors and adolescent obesity between rural and urban secondary school.
To identify the risk factors that influence the status of adolescent obesity between rural and urban secondary school.
RESEARCH QUESTIONS This study addresses the following research questions:
What are the percentage of adolescent who were obese and non-obese between rural and urban secondary school?
What are the relationships between the risk factors and adolescent obesity between rural and urban secondary school?
What are the risk factors that influence the status of adolescent obesity between rural and urban secondary school?
RESEARCH HYPOTHESISThis study focus on the following hypothesis:
To study the relationship between the risk factors and adolescent obesity between rural and urban secondary school
H1: There is significant relationship between the risk factors and adolescent obesity between rural and urban secondary school.
To identify the risk factors that influence the status of adolescent obesity between rural and urban secondary school.
H1: At least one of the coefficients on the parameters of the logistic regression modelling log(odds) of obesity as a function of log contribution and the risk factors are nonzero.
SIGNIFICANCE OF THE STUDYBased on the previous research that has been done before, there is the need to identify the risk factors and adolescent obesity between rural and urban secondary school for Form Four and Form Five. This is because many adolescent show a tendency to have obesity because of their lifestyle behavior. Besides that, this research is carried out to identify the percentage of adolescent who were obese and non-obese between rural and urban secondary school for Form Four and Form Five. Moreover, this research also interested to examine the relationship between the risk factors and adolescent obesity between rural and urban secondary school for Form Four and Form Five. Furthermore, this research would also like to investigate the possibility of risk factors that influence the status of obesity between rural and urban secondary school among adolescent. The knowledge of the risk factors would help in devising mechanisms of how to minimize or eliminating the risk factors of obesity. In addition, the findings of the study will also give direct information and benefit towards ourselves as well as other people. Other than that, it will also help the non-governmental organization (NGO) and international non-governmental organization (INGO) working in health sectors to plan a programmed related to healthy life style.
SCOPE OF THE STUDYThe scope of this study involved all of form four and form five students from secondary school that will be selected in Kota Bharu, Kelantan. The study will be carried out on all of Form Four and Form Five students from one of urban and rural secondary school that will be selected in Kota Bharu, Kelantan as the respondent.
LIMITATIONS OF THE STUDY
The limitations of this research are the population size of this research only consists for all Form Four and Form Five students from one of urban and rural secondary school in Kota Bharu, Kelantan.
CHAPTER TWO
LITERATURE REVIEWPREVIOUS STUDY
The commonness of obesity has reached worrying stages, influencing almost both developed and developing nations of all socio-economic groups, regardless of age, gender or race. Globally, approximately more than 22 million children below the age of 5 are extremely overweight and it has been founded that one in 10 children are overweight (Kosti & Panagiotakos, 2006). Technically, Malaysia is also one of the developing Asian countries that suffer from a fast widespread presence of adolescent obesity which is accordant with the global trends (Yang et al., 2017). Therefore, obesity does not happen overnight because it evolves slowly from poor eating habit and bad choices of lifestyle among individual. The result from Journal of Tropical Pediatrics research in 2016 shows that the occurrence of obesity and overweight increase continuously both in rural and urban secondary school (Zhang et al., 2016). However, the occurrence of obesity in urban secondary school are significantly higher compared with rural secondary school (Zhang, Wang, Zhao & Chu, 2016). The respondent of the particular research consist of students aged between 7 years old and 18 years old (Zhang et al., 2016).
National Health Service (NHS) in the United Kingdom also claim that childhood obesity is said can be a powerful sign of weight-related health issues that will affect their adulthood if they keep practicing the unhealthy lifestyle (National Health Service, 2016). Moreover, research in 2018 found that there is an association between obesity and the frequency of exercise among participant (Elsangedy et al., 2018). Moreover, in a recent article, the authors have stated that poor sleep and physical activity which is exercise can be included in a list of risk factor of obesity among adolescent (Ji et al., 2018). Furthermore, the Journal of Social Science & Medicine in
2017 declare that family income has a relationship with weight-related health issues which obesity among adolescent (Cook, Tseng, Tam, John & Lui, 2017). The previous authors in 2017 also claim that eating behavior affect the occurrence of obesity among adolescent (Demir & Bektas, 2017).
One of the strengths of this study is only a few studies regarding the prevalence of obesity among the adolescent were conducted in Kota Bharu, Kelantan (Yang et al., 2017). Hence, the researcher decided to do a deep research regarding the risk factors that might affect the obesity among adolescent. Thus, all these variables will be explained in detail in another part of the literature review of the previous journal and article as below.
PARENTS INCOMEIn 2016, there is a research that has been done to determine the factor that affects weight-related health problems among Asian adolescent with special attention towards Asian ethnicity, socioeconomic status (SES), and their interaction (Cook, Tseng, Bautista & John, 2016). The finding of the study indicates that there is an association between low family income and overweight that affect the Asian American adolescents unfavorably (Cook et al., 2016). At the same time, Alvarado’s journal in 2016 also mentioned that one of the time-varying factors which are income could affect the individuals’ weight status (Alvarado, 2016). However, the results of the research are not enough to support the statement. It is because the households’ income alone is not sufficient to be one of the risk factors of obesity among adolescent (Alvarado, 2016). Besides that, one of the International Association for the Study of Obesity articles stated that the socio-economic status (SES) indicator showed that the wealthy family was more likely suffer from obesity for both men and women (Dinsa, Goryakin, Fumagalli ; Suhrcke, 2012). Adolescent whose parents’ income was either low or average were more likely to be obese compared with adolescent whose family have a high income (Strauss & Knight, 1999).
EATING BEHAVIOR
The definition of eating behavior can be understood as a complex interplay of physiologic, psychological, social, and genetic factors that influence meal timing, quantity of food intake, and food preference (Grimm & Steinle, 2011). Next, eating behaviors can be one of the reason of the occurrence of obesity as well as metabolic syndrome and other complications that increased by a variety of common genetic variants (Grimm & Steinle, 2011). Adolescent with unpredictable characteristics might be associated with their obesity or overweight issues (Yavuz & Selcuk, 2018). Hence, the children with negative affectivity in infancy were suggested to have higher risk for weight-related health issues which is obesity (Yavuz & Selcuk, 2018). In fact, the behavior might affect their adulthood if they do not change it. In return, the parents who have children with negative temperamental affectivity were suggested to feed their children with the purpose of soothing rather than to please the satiety cues of the children which increases the likelihood of developing emotional eating (Yavuz & Selcuk, 2018). Furthermore, in another Journal of Eating Behaviors the factors that influence the adolescent obesity were found to be enjoyment of food, emotional overeating, food responsiveness, satiety responsiveness and also food fussiness (Demir & Bektas, 2017). However, satiety responsiveness is not used in this research. In addition, the previous research in 2017 found that one of the eating behaviors which is emotional feeding is the third most significant factor of adolescent weight-related problem which is obesity (Altan & Bektas, 2017).
EXERCISEExercise can be understand as activity requiring physical effort, carried out to sustain or improve health and fitness (Oxford Dictionary, n.d.). Next, exercise can decrease the fat stores in the body which has metabolic effects on the obesity as well as increasing caloric expenditure and improving glucose tolerance together with the lipid profile. As a result, exercise can reduces the damage of weight-related problems which is obesity (Paes, Marins & Andreazzi, 2015). The statement can be supported by referring the article in Korea that declared Trans-theoretical model based exercise (TTM) intervention combined with exercise classes were effective to control weight among obese adolescent (Ham et al., 2016). Furthermore, the increase of exercise
tolerance might also enable the adolescent to continued participate in any physical activity including school sports (Gow et al., 2016). This kind of physical activity can help to lose weight and enhanced the cardiometabolic outcomes (Gow et al., 2016). According to American Heart Association, cardiometabolic outcomes refer to obesity, hypertension, Type Two Diabetes Mellitus and also cardiovascular disease (Webb, S. R., & P., 2018). In addition, the result from the previous research done on medical student also showed that there are 54% of the students who do not exercise comparedwith students who do exercise with 46% respectively (Katuka et al., 2016). This show that the lack of exercise can cause obesity problems (Katuka et al., 2016).
SLEEPSleep restriction or short sleep can be describe as sleep time less than the average basal level of about 9 hour per night for adolescents (Carskadon, Acebo & Jenni, 2004). Available evidence suggests that disturbed sleep and restricted sleep are associated with de?cits in functioning across a wide range of indicators of psychological, interpersonal and somatic well-being (Roberts & Duong, 2014). Yet, the results in Journal of Psychosomatic Research revealed that there was no association between sleep restriction and obesity (Roberts & Duong, 2015). The authors of the journal also stated that sleep restriction did not increase future risk of obesity, nor did obesity increase risk of future sleep restriction (Roberts & Duong, 2015). When children were preschool-aged, approximately 25% had a bedtime of 8:00 p.m. or earlier, 50% had bedtimes after 8:00 p.m. but by 9:00 p.m. and 25% had a bedtime after 9:00 p.m (Anderson, Andridge & Whitaker, 2016). Earlier bedtimes at preschool age were associated with lower risk for adolescent obesity (Anderson, Andridge & Whitaker, 2016). The authors also mentioned that the earlier bedtime will results to lower mean of BMI and lower likelihood to have overweight issues (Anderson, Andridge & Whitaker, 2016). Sleep restriction has been regularly shown to increase hunger, appetite and food intake, with the increase in caloric intake in excess of the energy requirements of extended wakefulness (Reutrakul & Cauter, 2018). Meanwhile, additional sleep cause 14% reduction in overall appetite and 62% reduction in desire for sweet and salty foods (Reutrakul & Cauter, 2018). In another case, late bedtime was significantly associated with obesity whereas early wake-up was not (Sasaki et al., 2018).
CHAPTER THREE
METHODOLOGYINTRODUCTIONIn this chapter, the researchers describe and give an explanation about the method that will be used as an overview for this studies plan. This method is focusing only on the population of Rural and Urban Secondary Schools students for Form Four and Form Five in Kota Bharu, Kelantan.
STUDY POPULATIONThis study will carried out in Kota Bharu, Kelantan where the target population for this study is students from form four and form five of rural and urban secondary school. The population consists of 13138 students, which are then further divided into 41 schools.
3.2STUDY DESIGN
The study design will used is a cross-sectional design where the data is collected from the sample at a single point in time. The cross-sectional design is classified into a single cross-sectional design where only one sample is collected from the population and the information obtained is only once. Quantitative data analysis was employed to analyze the survey questionnaires. Using statistical methods, the results of the quantitative analysis can confirm or refute the hypotheses about the impact of characteristics of variables that contribute to the obesity. All data obtained from the questionnaires will be processed using IBM SPSS Statistics 24th Version. In this research, the inferential approach is applied where the survey questionnaires are conducted on the secondary school students to explore the factors that affect obesity among the adolescent of rural and urban secondary school for form four and form five.
3.3 DESCRIPTION OF SAMPLE3.3.1Included Criteria
The included criterion for this study is all Form Four and Form Five students from one of rural and urban secondary school that will be selected in Kota Bharu, Kelantan.
3.3.2 Excluded CriteriaThe excluded criteria are all other than Form Four and Form Five students from one of rural and urban secondary school that will be selected in Kota Bharu, Kelantan. Furthermore, all Form Four and Form Five absent students from one of rural and urban secondary school in Kota Bharu, Kelantan.
3.3.3Research Participants
The researcher chose Form Four and Form Five students from one of rural and urban secondary school. This is because one of the previous research articles found out that the pattern of period for changes in health-related behaviours started at the age of 16 years old (Pell et al., 2016). Moreover, there is also an articles discover adolescent at the age of 17 years old have excess weight compared with 15 years old adolescent (Mai & Wan, 2017).
3.3.4Body-Mass Index (BMI) Cut-Off Points
TABLE 3.1: Recommended Body-mass index (BMI) cut-off points for Asian populations
Body Weight Classification BMI cut-off points (kg/m2)
Underweight < 18.5
Normal Range 18.5 to 24.9
Overweight ? 25.0
Obese I 30.0 to 34.9
Obese II 35.0 to 39.9
Obese III ? 40.0
The researcher determine the classification of body weight by using WHO Expert Consultation (2004) as shown in the table. Body Mass Index (BMI) can be defined as body weight in kilogram (kg) divided by height (m2) (WHO Expert Consultation, 2004). BMI is important to all the individual because it can show them the degrees of underweight and overweight that might be related with non-communicable diseases.
3.4MEASURING INSTRUMENT
In this study, a questionnaire will used. A questionnaire is a structured set of questions that has been developed for obtaining information from the respondents. There are quite a lot of advantages when using the questionnaire in the study such as it is a less expensive way to reach more people, can avoid interviewer bias and the reliability of data collection. The questionnaire design is divided into four sections with 37 questions being asked. This questionnaire consist of open-ended and close-ended questions. The questionnaire also included the Part A: Demographic Information, Part B: Eating Behavior, Part C: Sleep, Part D: Exercises. The questions in the questionnaire
were adapted from several sources. The sources are from (Wardle, Guthrie, Sanderson, & Rapoport, 2001) and (Kautiainen, 2008).
The summary of the questionnaire is in the Table 3.1. The questionnaire that is used in this research has been obtained and adapted from the previous study that relates to this study. For this study, the questionnaire structured is adapted from several sources.TABLE 3.2: Questionnaire summarySection Construct No of item
A
B
C
D Demographic Information
Eating Behavior
Sleep
Exercises 8 questions
16 questions
7 questions
6 questions
3.4.1 Reliability of QuestionnaireThis two questionnaire can be used since the first questionnaire stated that majority of the Cohen’s kappa coefficient for test-retest reliability of the variable in the questionnaire were fair to good. According to Fleiss in 1981, kappa coefficients values between 0.40 and 0.75 are fair to good. Next, the authors of another questionnaire stated that the Cronbach’s alpha ranged from 0.74 to 0.91 were considered internally valid and good test-retest reliability (Wardle, Guthrie, Sanderson, & Rapoport, 2001). 3.5METHOD OF DATA COLLECTIONThe population of this study consists of the students of Form Four and Form Five in Kota Bharu, Kelantan who is currently from secondary school. The sample will choose from the rural and urban of secondary school. The researcher will uses direct questionnaire method. Direct questionnaire requires a set of prepared questions so that the data can be obtained systematically and accurately. In this method, the researcher will greet respondents and explain briefly his intention before giving the questionnaires to the respondents. The questionnaire consist of four sections. The first section of the
questionnaire consists of respondent’s demographic profiles such as gender, form, school area, name of the school, height, weight, do you think fall into the appropriate weight category for someone your height and gender and about parents total income in a month that influences obesity among the adolescent of rural and urban secondary school. The second section of the questionnaire that was highlighted was the eating behavior that influences obesity among the adolescent of rural and urban secondary school. The third section consists of questions regarding to sleep that influences obesity among the adolescent of rural and urban secondary school. The last section in thequestionnaire consists of several questions that are related with exercise which is one of the factors that influences obesity among the adolescent of rural and urban secondary school.
SAMPLING TECHNIQUESThe researcher will use a combination of stratified random sampling and cluster sampling technique to conduct this study. This study was conducted on Form Four and Form Five of rural and urban secondary schools students to determine specifically which factors that affect obesity among the adolescent. Stratified random sampling is used when the researcher wants to select a strata within the population (Thomas, 2006). This technique is beneficial to the research since it warrants the presence of the strata key within the sample. Stratified Sampling Techniques is a probability sampling technique where the researcher divides the entire population into different subgroups or strata. The population of secondary students in Kota Bharu, Kelantan is homogenous in terms of their age which is 16 and 17 year old. It is heterogeneous in terms of their school which are rural and urban. Next, cluster sampling is a technique that can be describe as the population is divided into clusters and then selected to be included in the sample (Jackson, 2015). Cluster sampling technique will be used to select one of rural and one of urban secondary schools by random sample. The first step for cluster sampling is, the researcher needs a sampling frame of the name of schools for rural and urban secondary school. Then, the researcher select one of rural and one of urban secondary schools by using random number that will be generated using a computer routine which is SPSS to determine which elements to be selected as the sample. After one of rural and one of urban secondary schools has been determine, all students Form Four and Form Five students from one of rural and
urban secondary school that will be selected as the sample. The selection of the one of urban and one of rural secondary school names is completely unbiased because the computer generates the number and these number correspond to the names on the list. The sample random sampling process is in the Figure 3.1. Furthermore, the researcher find the sample size by using Krejcie ; Morgan (1970) table.
576262585725Select one of rural and one of urban secondary schools by the random number
00Select one of rural and one of urban secondary schools by the random number
509587516471900069723001647825003105150161861500-17145038100Whole Population
All name form four and form five of secondary schools from Kota Bharu, Kelantan.
Sample one of rural and one of urban student from form four and form five secondary schools
Total = 375
Population form four and form five of secondary schools from Kota Bharu, Kelantan.
Total = 13138
00Whole Population
All name form four and form five of secondary schools from Kota Bharu, Kelantan.
Sample one of rural and one of urban student from form four and form five secondary schools
Total = 375
Population form four and form five of secondary schools from Kota Bharu, Kelantan.
Total = 13138
center95250Divides the entire population of students into different subgroups or strata which are rural and urban
00Divides the entire population of students into different subgroups or strata which are rural and urban
1095375160972500
FIGURE 3.1: Sampling process
SAMPLE SIZEThe researcher determines the sample size by using Krejcie and Morgan (1970) shows in table.
Based on the sample size table, the researcher chooses 375 samples from the 13138 population.
N= 13138 Students
n = 375 students
The populations consist of N objects.
The sample consists of n objects.
All possible samples of n objects are equally likely to occur.
TABLE 3.3: Margin Table
Next, researcher uses stratified sampling techniques to divides the entire population of students into different subgroups or strata which are rural and urban. Then, the researcher uses cluster sampling to select one of rural and one of urban secondary schools by the random number that will be generated using a computer routine which is SPSS to determine which elements are to be selected as the sample. After one of rural and urban secondary schools has been determine, the minimum of 381 students Form Four and Form Five students from one of rural and one of urban secondary school that will be selected as the sample.
THEORETICAL FRAMEWORKThe obesity among the adolescent of rural and urban secondary school are based on four factors which is parents’ income, eating behaviour, sleep and exercise. The theoretical framework for this research is shown in the Figure 3.2 below.
571580010002933701746250Eating behaviour
00Eating behaviour
281305168910Parents’ income
00Parents’ income
176657076962000
176530033464500
340042557150Obesity
00Obesity
2298702210435Exercise
00Exercise
236220805180Sleep
00Sleep
171323066675000171640571755000
FIGURE 3.2: The Theoretical Framework Relating the Four Variables in the Study
3.9PROCEDURE FOR DATA ANALYSIS3.9.1Descriptive StatisticsIn descriptive statistics, data are compiled, organized, summarized, and presented in suitable visual forms which are easy to understand (Sim & Wright, 2002). Thus, raw data are transformed into meaningful forms so that the user and manager can make conclusions just by taking a quick look at it (In & Lee, 2017). Hence, to determine whether the adolescent is obese and non-obese the BMI will be calculate by using their weight and height. In this research, the descriptive statistics will used are pie chart, bar chart and also histogram.3.9.2Inferential StatisticsIn inferential statistics, the sample is analyzed to make a generalization about the population (Banerjee & Chaudhury, 2010). If the sample is a good representation of the population, accurate conclusions about the population can be inferred from the analysis of this sample (Biau, Kernéis, & Porcher, 2008). This is because the sample values are a close representation of the actual values of the population of interest. Thus, inferential statistical techniques are used to make inferences about the population based on measurements obtained from the sample. Pearson chi-square test and Logistic regression model is used in the inferential statistics. 3.9.2.1Logistic RegressionThe Logistic regression is a procedure used to test model to predict categorical outcome with two or more categorical. The predictor variable can be either categorical or continuous, or a mix of both in one model. Here will use the procedure labelled Binary Logistic (Pallant, 2013).
Assumptions:
Logistic regression typically requires large sample size.
The predictor variable must be strongly related to dependent variable but not strongly related to each other. In short, it is said that there is no multicollinearity amount the independent variable.
The model must not have outlier.Pearson Chi-Square Test Chi-square test for independence is a procedure used to test when you have two categorical variable from a single population. It is used to determine whether there is significant association between the two variables (Coakes, 2013).
Assumption:
The sampling method is simple random sampling.
The variables under study are each categorical.
If sample data are displayed in a contingency table, the expected frequency count for each cell of the table is at least 5.
