Assignment Code: MFN-009/AST-4/TMA-4/2022-23

Course Code: MFN-009

Assignment Name: Research Methods and Biostatistics

Year: 2022-2023

Verification Status: Verified by Professor

Maximum Marks:100

This assignment is based on Units 1 -14 of the MFN-009 Course.

Section A - Descriptive Questions

There are eight questions in this part. Answer all questions.

1. a) Define research. Enlist the various scientific activities involved in a research process.

Ans) Research is a process of systematic inquiry that entails collection of data; documentation of critical information; and analysis and interpretation of that data/information, in accordance with suitable methodologies set by specific professional fields and academic disciplines.

The various scientific activities involved in a research process:

1. Stage I : Selection and Formulation of a Problem

2. Stage II : Formulation of Hypothesis

3. Stage III : Formulation of Research Design

4. Stage IV : Collection of Data

5. Stage V : Analysis and Interpretation of Data

6. Stage VI : Generalization

b) Define hypothesis. Enumerate basic characteristics of a good hypothesis. Enlist the three forms of Hypothesis.

Ans) A hypothesis is a proposed explanation for a phenomenon. For a hypothesis to be a scientific hypothesis, the scientific method requires that one can test it. Scientists generally base scientific hypotheses on previous observations that cannot satisfactorily be explained with the available scientific theories.

The basic characteristics of a good hypothesis are:

1. Hypotheses give research a direction and keep researchers from reading and collecting useless or too much data. They let you organise information based on both its importance and how it fits together.

2. Hypotheses should be written in a way that shows a difference or a relationship between the measurements used in the study.

3. The hypothesis should be clear and short. The statement should be a short summary of the expected relationship.

The three forms of Hypothesis are:

1. Declarative hypothesis

2. Null hypothesis

3. Hypothesis in question form

2. a) Explain the following with examples:

i) Cohort study

Ans) Cohort study, is an epidemiology study that observes a large group of people over a period of time. A group or groups of people/individuals are defined on the basis of presence or absence of exposure to a suspected risk factor for a disease and eligible participants are then followed-up over a period of time to assess the occurrence of the outcome.

ii) Randomised control trials

Ans) Randomized controlled trials (RCT) are prospective studies that measure the effectiveness of a new intervention or treatment. Although no study is likely on its own to prove causality, randomization reduces bias and provides a rigorous tool to examine cause-effect relationships between an intervention and outcome. This is because the act of randomization balances participant characteristics (both observed and unobserved) between the groups allowing attribution of any differences in outcome to the study intervention. This is not possible with any other study design.

iii) Case Control study

Ans) A case–control study is a type of observational study in which two existing groups differing in outcome are identified and compared on the basis of some supposed causal attribute. The starting point of most case control studies is the identification by researchers of an outcome or effect (e.g., lung cancer, heart disease, anaemia etc.) and a number of potential causative factors. If investigating lung cancer, for example, the factors selected might include smoking history, asbestos exposure etc.

b) What are the two sampling methods in research. Explain any two methods of probability sampling.

Ans) The two sampling methods in research are Probability or a Random Sampling and Non-probability Sampling.

The two methods of probability sampling are:

Stratified Sampling: Stratified random sampling takes into account how the main population is divided into a number of subpopulations that are all the same in at least one way (s). After this is done, it allows for a random selection of the needed number of units from each subpopulation. A sample can be chosen from each sub-population in any way, such as by chance or by some other method.

Cluster sampling is used when the population being studied is unlimited, when there isn't a list of population units, when the units are spread out geographically, or when it would be hard to sample each unit for administrative reasons. It involves putting the elementary units' population into groups or clusters that will be used as primary sampling units. The sample is then chosen from one or more of the clusters. So, in cluster sampling, the sampling unit is not a single person or thing in the population, but rather a group of people or things.

3. a) Differentiate between the following highlighting their uses in epidemiological studies:

i) Incidence and Prevalence

Ans) “Prevalence” to refer to the number of people currently diagnosed with a disease, and you’ll see “incidence” when referring to the new cases being diagnosed over a period of time. We need both measures to help assess the risk and burden of diseases on our community.

ii) Quota sample and snowball sample

Ans) Quota sampling may be used when a researcher is interested in investigating certain trait or characteristics of a certain subgroup. It also allows one to observe relationships between subgroups. Snowball sampling on the other hand is used where potential participants are hard to locate.

iii) Cross sectional and correlational study

Ans) Cross-sectional studies are considered a type of cohort study where only one comparison is made between exposed and unexposed subjects. They provide a snapshot of the outcome and the associated characteristics of the cohort at a specific point in time whereas Correlational studies aim to find out if there are differences in the characteristics of a population depending on whether or not its subjects have been exposed to an event of interest in the naturalistic setting.

b) Enlist the various factors affecting the choice of sampling method. Write down characteristics of a good sample.

Ans) The various factors affecting the choice of sampling method: The choice of sampling methods depends on several considerations unique to each individual project. These include issues related to the definition of population, availability of information about the structure of the population, the parameters to be estimated, the objectives of the analysis including the degree of precision required, and the availability of financial and other resources. This calls for appropriate selection of a sample for the conduct of any research study.

The characteristics of a good sample are:

1. representative of the population chosen

2. adequate and

3. accurate

4. a) Given below are the hemoglobin levels (gm %) recorded for 10 girls in class X.

9.8, 7.4, 10.3, 11.2, 12.5, 8.5, 9.0, 10.0, 10.5, 12.0

Calculation the mean, median, standard deviation and variance for the hemoglobin levels reported above.

Ans) 7.4, 8,5, 9, 9.8, 10, 10.3, 10.5, 11.2, 12, 12.5

Min= 7.4

Mean= 10.12

Median= 10.15

Max= 12.5

In statistics, a population is a set of similar items or events which pertains to a question or experiment.

x1= 9.8

x2= 7.4

x3= 10.3

x4= 11.2

x5= 12.5

x6= 8.5

x7= 9

x8= 10

x9= 10.5

x10= 12

b) Differentiate between two-tailed and one-tailed tests of significance.

Ans) The difference between two-tailed and one-tailed tests of significance:

1. In the one-tail, the alternate hypothesis has a single end (either left or right), but in the two-tail, the alternate hypothesis has two ends (both sides).

2. The region of Rejection (Critical Region) in the one-tail test lies either on the left or side of the probability distribution curve, while in two tail tests, the critical region lies on both sides.

3. The test parameter in the one-tail is either more or less than the critical value, while in the two-tail, the results are either within or outside the critical value.

4. The one-tail checks the relation between the variables in a single direction, while the two-tail checks in both directions.

5. a) Explain the concept of confidence interval and degree of freedom with the help of an example.

Ans) We can get an idea of what the population mean is by using the "central limit theorem" and the "normal probability curve." When we take a large random sample from the population to measure a variable and find the sample's mean, we can use the "central limit theorem" and the "normal probability curve" (M pop.). We can say that M is likely to be within 1.96 standard error units of M pop 95% of the time. In other words, there is a 95% chance that the mean of a random sample will be within 1.96 M units of M pop. There is also a 99% chance that the sample mean is within 2.58 M units of M pop. To be more specific, it can be said that there is a 95% chance that the limits M + 2.58 M surround the mean of the population with a 95% chance. The population mean is surrounded by these limits M + 1.96 M. The limits that surround the population mean are called "confidence intervals."

In the basic formula for standard deviation, we use N-1 instead of N in the denominator when working with small samples. If N is big enough, the difference between the two formulas may not seem like much. But when it comes to small samples, the "meaning" is different in a very important way. df stands for the "number of degrees of freedom," which is N-1. In a distribution, the "number of degrees of freedom" is the number of observations or values that don't depend on each other and can't be figured out by looking at the others. In other words, we can say that "degrees of freedom" mean the ability to change. To show why the df here is N – 1, we'll look at 5 scores: 5,6,7,8, and 9, where the mean is 7. This mean score will be used to estimate the average score of the whole population. The scores are –2, –1, 0, +1, and +2 points off from the average of 7. Mathematically, the sum of these deviations from the mean should be zero. Only four of the five deviations, or N – 1, can be chosen freely (independently), because the sum must be equal to zero for the fifth deviation to have a value greater than zero. With this condition, we can change four of the five deviations at random and fix the fifth. We could take the first four deviations to be –2, –1, 0, and +1. This would mean that the fifth deviation would have to be +2 for the total deviations to equal 0. In the same way, we can try other changes, and if the sum stays the same, one of the five deviations will be chosen automatically. So, only 4, i.e., (5 – 1), are free to change within the limits.

b) Test the difference in the attitude of male and female subjects using chi-square analysis towards exercise based on the distribution given here with (at 0.5 level of significance):

Ans) Actual Values:

8, 9

8, 5

Expected Values:

9.06667, 7.93333

6.93333, 6.06667

Chi-square Values:

0.12549, 0.143417

0.164103, 0.187546

Chi-square= 0.620556

Degrees of Freedom= 1

p= 0.430841

6. a) Enlist the characteristics of a good Interview. What are the advantages and limitations of using an interview in a research study?

Ans) The characteristics of a good Interview:

1. Avoid Yes/No Questions.

2. Avoid ”Multiple-choice“ and Double-barrelled Questions.

3. Don't Switch Topics Too Frequently.

4. Avoid the 'Why' Question.

5. Avoid Asking for Little Known Facts.

6. Avoid Imposing Concepts.

7. Avoid Leading Questions.

8. Listen Attentively.

The advantages of using an interview in a research study:

1. It provides flexibility to the interviewers

2. The interview has a better response rate than mailed questions, and the people who cannot read and write can also answer the questions.

3. The interviewer can judge the non-verbal behavior of the respondent.

4. The interviewer can decide the place for an interview in a private and silent place, unlike the ones conducted through emails which can have a completely different environment.

5. The interviewer can control over the order of the question, as in the questionnaire, and can judge the spontaneity of the respondent as well.

The limitations of using an interview in a research study:

1. Conducting interview studies can be very costly as well as very time-consuming.

2. An interview can cause biases. For example, the respondent’s answers can be affected by his reaction to the interviewer’s race, class, age or physical appearance.

3. Interview studies provide less anonymity, which is a big concern for many respondents.

4. There is a lack of accessibility to respondents (unlike conducting mailed questionnaire study) since the respondents can be in around any corner of the world or country.

   
   

b) Discuss briefly different types of attitude scales used in epidemiological studies.

Ans) Different types of attitude scales, which are quick and easy ways to measure attitudes, have been made possible by different scaling techniques. But the "equal-appearing intervals" (Thurstone Scales) and "summated ratings" (Likert Scales) methods have been used a lot in research on attitudes or opinions. When these techniques are used to make attitude scales, they are made up of a number of carefully edited and chosen items called "statements."

Thurstone and Chave were the first people to come up with the "equal-appearing intervals" method. This method gives a person's attitude score an absolute meaning based on the psychological continuum of scale value of each statement in the scale. If this score falls in the middle of the continuum of psychological states, the person's attitude is called "neutral." If it's closer to the good end of the continuum, it's called "favourable," and if it's closer to the bad end, it's called "unfavourable."

In the "method of summated ratings" that Likert came up with, the item score is found by giving statements that support a point of view weights of 5, 4, 3, 2, and 1 for strongly agree (SA), agree (A), undecided (U), disagree (D), and strongly disagree (SD). On the other hand, each answer to a statement that goes against this point of view is given a point value of 1, 2, 3, 4, or 5. On a certain attitude scale, a person's score is the sum of how he or she rates each item.

7. a) Give one mortality and one morbidity measure for assessing the health status of children in a community. Indicate how will u measure them.

Ans) Death is easy to identify in nearly all the cases and the date of death is generally available in records. Thus, the mortality statistics are considered reliable and used all across the world. A higher rate of mortality in children is considered an indicator of poor health, though this may not be so for old age. Indicators used to measure deaths in a population are crude death rate (CDR), child morality rate, life expectancy at birth, etc. Among the child mortality indicators, the infant mortality rate is widely used as an indicator of health status and development. There are many other indicators that measure mortality in children beginning with conception.

Morbidity is departure from health. This results or has potential to result in at least some restriction in performing the normal activities of life. Morbidity in children could be in terms of infectious diseases such as diarrhoea, pneumonia and tetanus, or chronic such as congenital anomalies and thalassemia. The magnitude of morbidity in children can be measured by: (i) the number of children affected, (ii) the number of episodes or spells of sickness, particularly for acute conditions, (iii) the duration of illness, and (iv) and severity of illness. Morbidity is not so easy to quantify as mortality.

b) Learners (n=10) enrolled in M. Sc . DFSM, scored following marks (out of 100) in theory (X) and practical (Y). Find out the correlation coefficient and level of significance for the two variables.

X: 46, 69, 79, 47, 35, 57, 98, 69, 58, 79

Y: 40, 50, 54, 47, 57, 40, 49, 46, 44, 48

Ans) Number of value pairs: 10

Covariance x and y: 10.25

Standard deviation x: 17.87

Standard deviation y: 5.18

Correlation coefficient: 0.11

Test variable: 2.85

Degrees of freedom: 8

95% significance: yes

99% significance: no

The formulas are:

n= number of value pairs

∑= sum i=1 to n

xm= mean of all xi

ym= mean of all yi

Covariance x and y: sxy = 1/n * ∑(xi-xm)(yi-ym)

Standard deviation x: sx =

Standard deviation y: sy =

Correlation coefficient: rxy = sxy / (sx * sy)

Test variable: t=

Degrees of freedom: df= n-2

t-distribution

8. a) The nutrition score of males (n=10)and female (n=12) students of M.Sc. (DFSM) programme is given herewith:

Males:10, 12, 13, 10, 14, 15, 10, 12, 11,15

Females:08, 12, 16, 14, 15, 12, 13, 15, 08, 13, 10, 09

Based on the data given herewith:

i) Calculate the standard error of the difference between the means of the two groups.

ii) Test the significance of the difference between the mean nutrition score of males and females at 0.05 level of significance.

Ans) Males:

Count numbers= 10

Sum= 122

Mean= 12.2

Standard deviation= 1.988858

Standard variance= 3.955556

Population deviation= 1.886796

Population variance= 3.560000

Female:

Count numbers= 12

Sum= 145

Mean= 12.083333

Standard deviation= 2.778434

Standard variance= 7.719697

Population deviation= 2.660148

Population variance= 7.076389

Relative Difference [(B-A)/A] = -0.009563

Observed Difference vs. Error Distribution under H0

b) What are the characteristics of case study. Explain them.

Ans) The characteristics of case study are:

Continuity in Investigation: It is important to keep looking into situations for a long time, until the underlying causes are found and plausible patterns of how they interact with each other are found. For example, the problems that caused a nutrition education programme to fail can't be looked at all at once. A researcher may have to ask questions for a long time.

Completeness: A good case study involves gathering a lot of information about both the inside and outside of the unit being studied. The process of collecting data will keep going until the data is complete and a full picture of the unit emerges.

Authenticity of Data: A case study report must be based on information that is meaningful, valid, and reliable about the case. Case studies are a good place to use both qualitative and quantitative methods, such as interviews, observations, record surveys, and test questionnaires. Using multiple methods to collect data and cross-checking data using different methods can make sure that the data is accurate.

Confidential Recording: All of the necessary information, such as the relationships between anganwadi workers and the people they help, confidential records, documents about the institution, etc., must be handled with care and kept secret.

Intellectual Synthesis: Since a case study uses more than one method of research and looks at all of the important things that have happened to the unit, the data must be put together in the right way to show how unique the unit is and how important relationships are. A good case study can be done by a skilled investigator who is well-versed in theory, has a good sense of things, and can write well.

Section B - OTQ (Objective Type Questions)

2. a) Define the following:

i) Epidemiology

Ans) Epidemiology is the study and analysis of the distribution, patterns and determinants of health and disease conditions in a defined population. It is a cornerstone of public health, and shapes policy decisions and evidence-based practice by identifying risk factors for disease and targets for preventive healthcare.

ii) Plagiarism

Ans) Plagiarism is the fraudulent representation of another person's language, thoughts, ideas, or expressions as one's own original work. While precise definitions vary, depending on the institution, such representations are generally considered to violate academic integrity and journalistic ethics as well as social norms of learning, teaching, research, fairness, respect and responsibility in many cultures. It is subject to sanctions such as penalties, suspension, expulsion from school or work, substantial fines and even imprisonment

iii) Bias

Ans) In statistics, bias is a prejudice in a general or specific sense, usually in the sense for having a preference to one particular sample, perspective, external influence etc.

iv) Quasi Experimental Design

Ans) Quasi-experiments are studies that aim to evaluate interventions but that do not use randomization. Similar to randomized trials, quasi-experiments aim to demonstrate causality between an intervention and an outcome. Quasi-experimental studies can use both preintervention and postintervention measurements as well as nonrandomly selected control groups.

v) Statistics

Ans) Statistics is the discipline that concerns the collection, organization, analysis, interpretation, and presentation of data. In applying statistics to a scientific, industrial, or social problem, it is conventional to begin with a statistical population or a statistical model to be studied. Populations can be diverse groups of people or objects such as "all people living in a country" or "every atom composing a crystal". Statistics deals with every aspect of data, including the planning of data collection in terms of the design of surveys and experiments.

vi) Proximity error

Ans) Proximity error occurs when, due to the ordering, or polarity, of the differential scales, one answer on the semantic differential results in another answer to a subsequent question being substantially changed from what it would otherwise be. A computer produced a set of semantic differential questionnaires which were controlled for various kinds of proximity error--effects due to order of concept presentation, of adjective presentation, and of order of adjectives within a particular scale. Three experiments were conducted varying questionnaires and types of proximity error. In each experiment all measures indicated no significant differences in response traceable to questionnaire format manipulations, showing that proximity error was not a problem in administering semantic differential questionnaires

vii) Structured Questionnaire

Ans) Structured questionnaire is a document that consists of a set of standardized questions with a fixed scheme, which specifies the exact wording and order of the questions, for gathering information from respondents.

viii) Documents

Ans) Documents are records which normally come to the researcher ‘ready-made’. They may describe a process of personal/group development, or the occurrence of an event/ disease. Some other people, either a participant in a social situation or process, or the originator of a system of recording, has already determined the form/type of data. These data are reviewed in terms of the research problem before they are actually used by the researcher.

ix) Percentile Rank

Ans) A percentile rank indicates how well a student performed in comparison to the students in the specific norm group, for example, in the same grade and subject. A student's percentile rank indicates that the student scored as well as, or better than, the percent of students in the norm group

x) Non parametric test

Ans) Nonparametric statistics is the branch of statistics that is not based solely on parametrized families of probability distributions. Nonparametric statistics is based on either being distribution-free or having a specified distribution but with the distribution's parameters unspecified.

3. Differentiate between the following:

i) Null hypothesis and Alternative hypothesis

Ans) The difference between Null hypothesis and Alternative hypothesis:

1. A null hypothesis is a statement, in which there is no relationship between two variables. An alternative hypothesis is a statement; that is simply the inverse of the null hypothesis, i.e., there is some statistical significance between two measured phenomenon.

2. A null hypothesis is what, the researcher tries to disprove whereas an alternative hypothesis is what the researcher wants to prove.

3. A null hypothesis represents, no observed effect whereas an alternative hypothesis reflects, some observed effect.

4. If the null hypothesis is accepted, no changes will be made in the opinions or actions. Conversely, if the alternative hypothesis is accepted, it will result in the changes in the opinions or actions.

5. As null hypothesis refers to population parameter, the testing is indirect and implicit. On the other hand, the alternative hypothesis indicates sample statistic, wherein, the testing is direct and explicit.

6. A null hypothesis is labelled as H0 (H-zero) while an alternative hypothesis is represented by H1 (H-one).

7. The mathematical formulation of a null hypothesis is an equal sign but for an alternative hypothesis is not equal to sign.

8. In null hypothesis, the observations are the outcome of chance whereas, in the case of the alternative hypothesis, the observations are an outcome of real effect.

ii) Type I error and Type II error

Ans) The difference between Type I error and Type II error:

1. Type I error is an error that takes place when the outcome is a rejection of null hypothesis which is, in fact, true. Type II error occurs when the sample results in the acceptance of null hypothesis, which is actually false.

2. Type I error or otherwise known as false positives, in essence, the positive result is equivalent to the refusal of the null hypothesis. In contrast, Type II error is also known as false negatives, i.e., negative result, leads to the acceptance of the null hypothesis.

3. When the null hypothesis is true but mistakenly rejected, it is type I error. As against this, when the null hypothesis is false but erroneously accepted, it is type II error.

4. Type I error tends to assert something that is not really present, i.e., it is a false hit. On the contrary, type II error fails in identifying something, that is present, i.e., it is a miss.

5. The probability of committing type I error is the sample as the level of significance. Conversely, the likelihood of committing type II error is same as the power of the test.

6. Greek letter ‘α’ indicates type I error. Unlike, type II error which is denoted by Greek letter ‘β’.

iii) Odds ratio and risk ratio

Ans) The difference between Odds ratio and risk ratio:

iv) t-Test and paired t-Test

Ans) The difference between t-Test and paired t-Test:

1.  

2. Two-sample t-test is used when the data of two samples are statistically independent, while the paired t-test is used when data is in the form of matched pairs. There are also some technical differences between them. To use the two-sample t-test, we need to assume that the data from both samples are normally distributed, and they have the same variances.

3. For paired t-test, we only require that the difference of each pair is normally distributed. An important parameter in the t-distribution is the degrees of freedom. For two independent samples with equal sample size n, df = 2(n-1) for the two-sample t-test. However, if we have n matched pairs, the actual sample size is n (pairs) although we may have data from 2n different subjects.

v) Reliability and Validity

Ans) The difference between Reliability and Validity:

1. The degree to which the scale gauges, what it is designed to gauge, is known as validity. On the other hand, reliability refers to the degree of reproducibility of the results, if repeated measurements are done.

2. When it comes to the instrument, a valid instrument is always reliable, but the reverse is not true, i.e., a reliable instrument need not be a valid instrument.

3. While evaluating multi-item scale, validity is considered more valuable in comparison to reliability.

4. One can easily assess the reliability of the measuring instrument, however, to assess validity is difficult.

5. Validity focuses on accuracy, i.e., it checks whether the scale produces expected results or not. Conversely, reliability concentrates on precision, which measures the extent to which scale produces consistent outcomes.