Assignment Code: MFN-009/AST-4/TMA-4/2023-24

Course Code: MFN-009

Assignment Name: Research Methods and Biostatistics

Year: 2023-2024

Verification Status: Verified by Professor

Section A - Descriptive Questions

Q1a) What do you understand by the term ‘Nutritional Epidemiology’? Explain the descriptive variables related to community health used in research.

Ans) Nutritional epidemiology is a subfield of epidemiology that focuses on the study of dietary patterns and their effects on human health and disease in populations. It examines the relationship between nutrition, health, and diseases through the collection and analysis of dietary data, anthropometric measurements, and various health-related outcomes.

Descriptive variables related to community health used in research often include

Demographic Variables: These variables include age, gender, ethnicity, socioeconomic status, and other population characteristics. They are crucial for identifying potential disparities and understanding the distribution of health-related outcomes within a community.

Dietary Variables: These encompass dietary habits, food consumption patterns, nutrient intake, and dietary quality. Common dietary variables include the consumption of specific food groups, nutrients, dietary patterns, and adherence to dietary guidelines.

Anthropometric Variables: These variables involve physical measurements of individuals within a community, such as height, weight, body mass index (BMI), waist circumference, and body fat percentage. These measurements are used to assess body composition and nutritional status.

Clinical and Biochemical Variables: These include health indicators such as blood pressure, cholesterol levels, blood glucose, and other clinical or biochemical markers. They are essential for evaluating the association between diet and health outcomes.

Health Outcomes: These variables focus on specific health conditions, diseases, or health-related events within the community. Examples include the prevalence of chronic diseases (e.g., diabetes, cardiovascular disease), mortality rates, and incidence of specific health events.

Lifestyle Variables: Factors such as physical activity, smoking status, alcohol consumption, and other lifestyle behaviours may be considered when assessing the relationship between nutrition and health.

Q1b) What is null hypothesis? What are the characteristics of a good hypothesis?

Ans) The null hypothesis (H0) is a fundamental concept in hypothesis testing, especially in statistical analysis. It represents the hypothesis that there is no significant difference, relationship, or effect in the data or population under investigation.

Characteristics of a good null hypothesis include

1. Testability: A good null hypothesis should be testable, meaning that it can be subjected to empirical evaluation through data collection and statistical analysis.

2. Specificity: The null hypothesis should be specific and well-defined. It should state exactly what is being tested or compared.

3. Falsifiability: A valid null hypothesis must be falsifiable, meaning it can be proven false if evidence supports an alternative hypothesis (the research hypothesis).

4. Consistency: The null hypothesis should be consistent with current knowledge and theory. It should not contradict established facts or principles.

5. Independence: The null hypothesis should be independent of the research hypothesis. It should stand alone and not be based on or influenced by the expected outcome of the study.

6. Simple: The null hypothesis should be simple and straightforward, avoiding unnecessary complexity. It should focus on a single, specific relationship or comparison.

7. Measurable: The null hypothesis should involve variables and relationships that can be measured or quantified.

8. Negative Statement: The null hypothesis typically states the absence of an effect, relationship, or difference. It is often formulated as "there is no difference" or "there is no association."

Q2a) What is a cross-sectional descriptive study? Explain with example. List the strengths and limitations of descriptive cross-sectional study.

Ans) Definition: A cross-sectional descriptive study is a type of observational research design that involves collecting data from a population or sample at a single point in time or over a short period. It aims to describe the prevalence of various characteristics, conditions, or exposures within the study population.

Example: Suppose a research study is conducted to assess the dietary habits and nutritional status of a sample of 500 adults in a particular region during the year 2022. Researchers collect data on food consumption patterns, nutrient intake, and body mass index (BMI) for each participant at a specific point in time.

Strengths of Descriptive Cross-Sectional Studies

1. Efficiency: They are relatively quick and cost-effective to conduct, making them suitable for studying large populations.

2. Snapshot of Prevalence: They provide a snapshot of the prevalence of characteristics or conditions in the population.

3. Useful for Hypothesis Generation: They can be valuable for generating hypotheses for further research.

4. Limitations of Descriptive Cross-Sectional Studies

5. No Causality: They cannot establish cause-and-effect relationships; they can only describe associations.

6. Temporal Relationship: They do not provide information on the temporal sequence of events.

7. Selection Bias: There may be selection bias, as participation is limited to individuals available at the time of the study.

8. Limited for Rare Outcomes: They are less suitable for studying rare conditions or exposures.

   
   

Q2b) Explain the following with examples:

i) Experimental Research

Ans) Definition: Experimental research is a type of research design where researchers manipulate one or more independent variables to observe their impact on dependent variables. It involves the random assignment of participants to experimental and control groups.

Example: An experiment is conducted to investigate the effect of a new drug on reducing blood pressure. Participants are randomly assigned to either the treatment group (receives the new drug) or the control group (receives a placebo). Blood pressure measurements are compared between the two groups.

ii)Correlation study

Ans) Definition: A correlation study examines the relationships between two or more variables without experimental manipulation. It measures the strength and direction of associations between variables.

Example: A study examines the correlation between hours of study (independent variable) and students' exam scores (dependent variable) among a group of high school students. It finds a positive correlation, suggesting that more hours of study are associated with higher exam scores.

iii) Case control study.

Ans) Definition: A case-control study is an observational study design that compares individuals with a specific outcome or condition (cases) to individuals without that outcome (controls). It aims to identify factors associated with the development of the condition.

Example: In a case-control study on lung cancer, researchers select a group of individuals diagnosed with lung cancer (cases) and a group without lung cancer (controls). They collect data on various exposures (e.g., smoking history, occupational exposure) and analyse the differences between the two groups to identify potential risk factors for lung cancer.

Q3a) How is purposive sampling different from incidental sampling? Discuss the characteristics of a good sample.

Ans) Comparison between Purposive sampling and Incidental sampling:

Characteristics of a Good Sample

1. Representativeness: A good sample should represent the population from which it is drawn. It should include a variety of characteristics and demographics that mirror the population as closely as possible.

2. Adequate Size: The sample size should be sufficiently large to provide meaningful results and statistical power while being practical to manage.

3. Randomness: In probability sampling methods, randomness is essential to ensure that each member of the population has an equal chance of being selected.

4. Unbiased: The selection process should not introduce biases that affect the results. It should be free from favouritism or systematic errors.

5. Accuracy: The sample should provide accurate data that can be used to draw valid conclusions about the population.

6. Feasibility: The sample should be feasible in terms of time, resources, and logistics. It should be practical to collect data from the selected participants.

Q3b) What are the different types of scales used in behavioural sciences for data measurement ?

Ans) Types of Scales in Behavioural Sciences:

1. Nominal Scale: This scale classifies data into categories or groups without any inherent order or ranking. It is used for categorical variables. Example: Gender (categories: male, female, non-binary).

2. Ordinal Scale: The ordinal scale arranges data into ordered categories or ranks, but the intervals between categories are not consistent. It indicates relative differences but not the magnitude of differences. Example: Educational levels (categories: high school, bachelor's degree, master's degree).

3. Interval Scale: The interval scale has ordered categories with consistent intervals between them. It lacks a true zero point, and differences can be measured. Example: Temperature (measured in degrees Fahrenheit or Celsius).

4. Ratio Scale: The ratio scale has ordered categories with consistent intervals and a true zero point. It allows meaningful ratios and comparisons. Example: Age, income, weight.

Q3c) What do you understand by validity and reliability of a research tool?

Ans) Validity: Validity refers to the extent to which a research tool measures what it is intended to measure. It assesses whether the tool is capable of accurately capturing the concept or construct under investigation. Validity includes content validity, criterion validity, and construct validity.

Reliability: Reliability measures the consistency and stability of a research tool's results over time and across different situations. A reliable tool produces similar results with repeated measurements. Types of reliability include test-retest reliability, internal consistency, and inter-rater reliability.

Q4a) What is an attitude scale? Enlist its uses and limitations.

Ans) Attitude Scale: An attitude scale is a research tool designed to measure individuals' attitudes, opinions, or perceptions regarding specific topics, objects, or issues. These scales are used to quantify and assess the intensity and direction of attitudes. Attitude scales can be structured in various ways, including Likert scales, semantic differential scales, and Thurstone scales.

The uses and limitations of attitude scales are as follows:

Uses

Research: Attitude scales are commonly used in social and behavioural research to investigate people's opinions and preferences, helping researchers understand their attitudes toward a particular subject.

Market Research: Businesses and marketers use attitude scales to assess consumer preferences and perceptions of products and services, aiding in product development and marketing strategies.

Evaluation: Attitude scales are employed to evaluate the effectiveness of interventions, programs, or campaigns. For instance, assessing attitudes toward a public health campaign.

Psychological Assessment: Attitude scales are utilized in psychology to evaluate individuals' beliefs, feelings, and attitudes to aid in diagnosis and treatment planning.

Limitations

Response Bias: Participants may provide socially desirable responses or may not be entirely truthful, leading to response bias.

Validity: The validity of attitude scales can be affected by the accuracy of questions and the construct being measured.

Sensitivity: Some individuals may find the response options too limiting, and this can lead to a lack of sensitivity in measuring nuanced attitudes.

Social Desirability Bias: Respondents may answer in a way they believe is socially acceptable rather than expressing their true attitudes, impacting accuracy.

Q4b) Explain the various types of observations. Discuss the stages in the process of observation.

Ans) Types of Observations:

1. Structured Observations: In structured observations, the researcher uses a predetermined set of categories or codes to record specific behaviours, making the data more quantifiable and objective.

2. Unstructured Observations: Unstructured observations involve more open-ended, flexible data collection without predefined categories, allowing researchers to capture a broader range of behaviours.

3. Participant Observations: Researchers immerse themselves in the environment being observed, actively participating, and recording their observations.

Stages in the Process of Observation

1. Preparation: Define the research objectives, select the observation method, and identify the target population.

2. Data Collection: Conduct the actual observations, record data, and remain unobtrusive to avoid influencing the subjects.

3. Recording Data: Use clear and concise notation or technology (e.g., video recording) to record observations systematically and accurately.

4. Data Analysis: Organize and analyse the data, looking for patterns, themes, and trends. This may involve coding and categorizing observed behaviours.

5. Interpretation: Draw conclusions from the observed data, consider the implications of the findings, and relate them to the research objectives.

6. Reporting: Present the observational findings in a clear and meaningful manner, often using descriptive narratives or quantitative summaries.

Q4c) Briefly describe the methods of data collection.

Ans) Methods of Data Collection:

1. Surveys: Structured questionnaires or interviews to gather information from participants.

2. Observation: Systematically watching and recording behaviour or events.

3. Interviews: Conversations with individuals or groups to obtain qualitative data.

4. Experiments: Controlled investigations to assess the impact of variables on outcomes.

5. Document Analysis: Examining existing documents, records, or archives for information.

6. Focus Groups: Group discussions to explore participants' attitudes and experiences.

7. Content Analysis: Analyzing the content of written or visual materials.

8. Case Studies: In-depth examinations of specific individuals, organizations, or situations.

Q5a)Enumerate the graphs you will use for representing frequency distribution?

Ans) Frequency distributions can be visually represented using various types of graphs, each of which is suitable for different types of data.

Common graphs used for representing frequency distributions include:

1. Histogram: A histogram displays the distribution of continuous data by dividing it into intervals (bins) on the x-axis and representing the frequency of data points within each interval using bars. It provides a visual representation of the data's shape and central tendencies.

2. Bar Chart: Bar charts are similar to histograms but are used for discrete or categorical data. Each category is displayed as a separate bar, and the height of each bar represents the frequency of that category.

3. Frequency Polygon: A frequency polygon is a line graph created by connecting the midpoints of the bars in a histogram or the points representing frequencies in a bar chart. It is used to visualize the distribution of continuous data.

4. Ogive (Cumulative Frequency Curve): An ogive is a line graph that represents the cumulative frequency distribution. It shows how many data points fall below or within certain values. There can be a less than ogive (cumulative frequency) and a greater than ogive (cumulative frequency greater than).

5. Pie Chart: A pie chart is used to represent parts of a whole or the distribution of a single categorical variable. It divides the circle into slices, with each slice representing a category, and the size of the slice corresponds to the proportion of data points in that category.

6. Box-and-Whisker Plot: A box-and-whisker plot displays the distribution of data by showing the median, quartiles, and potential outliers. It is particularly useful for identifying the spread and central tendencies of data.

Q5b)Describe the characteristics of a Normal Probability Distribution.

Ans) Characteristics of a Normal Probability Distribution:

1. Symmetry: The normal distribution is symmetric, with the mean, median, and mode all being equal and located at the center of the distribution.

2. Bell-Shaped Curve: The distribution takes the shape of a bell curve, with the majority of data clustered around the mean, and fewer data points in the tails.

3. Mean, Median, and Mode: They are all equal and located at the center of the distribution.

4. Constant Standard Deviation: The spread of data is consistent, and the standard deviation is the same for all values.

5. Tail Extends to Infinity: The tails of the distribution extend infinitely in both directions.

6. Empirical Rule: About 68% of data falls within one standard deviation from the mean, approximately 95% within two standard deviations, and nearly 99.7% within three standard deviations.

7. Asymptotic: The tails of the distribution approach but never touch the x-axis.

8. Summation of Probabilities: The area under the curve is equal to 1, indicating that the probabilities of all possible outcomes add up to 100%.

Q5c)What do you understand by the term Relative risk and Odd’s ratio?

Ans) Relative Risk (RR): Relative risk is a measure used in epidemiology to quantify the risk of an event or outcome occurring in one group compared to another. It is the ratio of the probability of the event occurring in the exposed group to the probability of the event occurring in the non-exposed (control) group.

The formula for relative risk is:

RR= Risk in the Non-Exposed Group

Risk in the Exposed Group

Relative risk values greater than 1 indicate an increased risk in the exposed group, while values less than 1 suggest a reduced risk.

Odds Ratio (OR): The odds ratio is another measure used to assess the relationship between an exposure and an outcome. It is the ratio of the odds of the event occurring in the exposed group to the odds of the event occurring in the non-exposed group.

The formula for odds ratio is:

Odds ratios are commonly used in case-control studies. An odds ratio equal to 1 indicates that there is no difference in odds between the exposed and non-exposed groups. Values greater than 1 suggest an increased odds, while values less than 1 indicate a decreased odds of the event.

Q6a) Ten scores of 10 learners enrolled in MFN-009 is given below:

40,36,29,14,27,45,50,12,39,19

Calculate the mean, median, standard deviation, and variance for the above data.

Ans)

Mean (Average)

Mean (μ) = (Σx) / n

where Σx represents the sum of all data points, and n is the number of data points.

For the provided data:

Σx = 40 + 36 + 29 + 14 + 27 + 45 + 50 + 12 + 39 + 19 = 291

n = 10

Mean (μ) = 291 / 10 = 29.1

Median

The median is the middle value when data is arranged in ascending order. Since there are 10 data points, the median will be the average of the 5th and 6th values (when sorted).

First, sort the data:

12, 14, 19, 27, 29, 36, 39, 40, 45, 50

Median = (29 + 36) / 2 = 65 / 2 = 32.5

Standard Deviation

The formula to calculate the standard deviation (σ) is:

[σ = √[(Σ(xi - μ)²) / (n - 1)]

where Σ represents the summation symbol, xi is each data point, μ is the mean, and n is the number of data points.

First, find the squared differences from the mean (μ) for each data point:

Variance

Variance (σ²) is the square of the standard deviation:

σ 2 =(13.27) 2 ≈175.87

Q6b) Describe the assumptions on which non-parametric tests are based?

Ans) Assumptions on which non-parametric tests are based:

1. Random Sampling: Data should be obtained through random sampling to ensure that the sample is representative of the population.

2. Independence: Data points should be independent of each other. The value of one data point should not be influenced by the value of another data point.

3. Categorical Data: Non-parametric tests are typically used with categorical or ordinal data, as they do not assume a specific underlying distribution of the data.

4. Mutually Exclusive Categories: Data should be divided into mutually exclusive categories or groups.

5. Homogeneity of Variance: Non-parametric tests do not assume that groups have equal variances.

6. Scale of Measurement: Non-parametric tests are applicable for nominal, ordinal, and sometimes interval data, but not typically for ratio data.

Q7a) In a sample of 100 children, 1-3 years of age, mean (SD) intake of calcium = 175 (5.82) mg. Compute the standard error of mean.

Ans) To compute the standard error of the mean (SEM), you can use the following formula:

SEM = SD / sqrt(n)

Where:

SD is the standard deviation of the sample.

n is the sample size.

Given:

SD = 5.82 mg

n = 100

Let's plug in the values:

SEM = 5.82 / sqrt(100)

Now, calculate the square root of 100, which is 10:

SEM = 5.82 / 10

SEM = 0.582 mg

So, the standard error of the mean is 0.582 mg.

Q7b) The theory and practical marks of 10 students are given below:

Theory (x): 24 34 33 22 37 23 38 33 44 35

Theory (y): 15 21 28 31 18 24 36 32 27 18

Find the correlation coefficient between x and y.

Ans) To find the correlation coefficient between two sets of data (in this case, theory marks x and y), we can use the following formula for the Pearson correlation coefficient:

Let's calculate step by step.

Given data:

x:24,34,33,22,37,23,38,33,44,35

y:15,21,28,31,18,24,36,32,27,18

Step 1: Calculate the mean of x and y.

After calculating the correlation coefficient using the formula, we get:

R ≈ −0.198

Therefore, the correlation coefficient between x and y is approximately -0.198.

Q8a) What is NUDIST software? Discuss the features which are available with this software.

Ans) NUDIST (Non-numerical Unstructured Data Indexing, Searching, and Theorizing) is a software program used for managing and analyzing qualitative data. It's designed for researchers in various fields, including social sciences, to assist in systematically analyzing large volumes of text or other qualitative data sources.

Features available with NUDIST software

1. Data Management: NUDIST allows users to import, organize, and manage various forms of qualitative data, such as text documents, audio files, images, and video. It provides a structured platform to store and access your data.

2. Coding: One of the primary features is the ability to code data. Researchers can create codes or categories to label and categorize segments of text or other qualitative data. These codes help in identifying patterns, themes, and relationships within the data.

3. Query and Retrieval: NUDIST enables users to search and retrieve data based on specific criteria or codes. Researchers can quickly locate relevant data segments within the vast dataset.

4. Visualization: The software often includes tools for creating visual representations of coded data, such as concept maps, charts, and diagrams. These visual aids help in understanding the data and communicating findings.

5. Integration: NUDIST software typically integrates with various qualitative research methods and methodologies, making it versatile for different research needs.

6. Data Sharing: It allows researchers to collaborate and share data and findings with others, which is especially useful for team-based research.

7. Advanced Analysis: NUDIST can provide advanced qualitative analysis features, including text mining, sentiment analysis, and inter-rater reliability checks for coding consistency.

Q8b) Discuss the methods used for analysis of qualitative data?

Ans) Qualitative data analysis involves systematically examining non-numeric data, such as text, audio, or visual content, to identify patterns, themes, and insights. Common methods for analyzing qualitative data include:

1. Thematic Analysis: Researchers identify recurring themes or patterns in the data. This method involves coding the data, creating categories, and organizing content into themes. It's widely used for understanding participants' perspectives and experiences.

2. Content Analysis: Researchers analyse the content of the data, often using predefined categories. This method is useful for summarizing and describing large volumes of textual data.

3. Grounded Theory: A systematic and iterative approach to develop theories or conceptual frameworks based on the data. Researchers identify concepts and categories that emerge from the data itself, without preconceived notions.

4. Narrative Analysis: Focusing on the narratives and stories within the data, this method examines the structure, content, and meaning of narratives to understand individual experiences.

5. Discourse Analysis: Analyzing language, communication, and discourse patterns in the data. Researchers examine how language constructs meaning, power, and social contexts.

6. Framework Analysis: Involves structured data management and categorization using matrices and charts. It's useful for policy and case-based analysis.

7. Constant Comparative Analysis: A method commonly used in grounded theory, involving the ongoing comparison of data segments to identify similarities and differences.

8. Visual Analysis: Applicable when analyzing visual data like photographs, videos, or artworks. Researchers interpret and code visual elements and their meaning.

Section B - OTQ (Objective Type Questions)

Q1) Define the following:

i) Biostatistics

Ans) Biostatistics is the application of statistical methods to analyse and interpret data related to biological, health, and medical phenomena. It involves the collection, organization, analysis, and interpretation of data to make informed decisions and draw conclusions in the fields of biology and health sciences.

ii) Plagiarism

Ans) Plagiarism refers to the act of using someone else's words, ideas, or work without giving them proper credit, often presenting it as one's own. It is a form of academic or intellectual dishonesty and is considered unethical and unacceptable in research, writing, and other creative endeavours.

iii) Double blind trial

Ans) A double-blind trial is a research study or clinical trial in which both the participants and the researchers are unaware of who is receiving the treatment and who is receiving a placebo or control. This helps eliminate biases and ensures the validity of the results.

iv) Halo effect

Ans) The halo effect is a cognitive bias in which a person's overall impression of someone or something influences their perception of that entity's specific qualities or characteristics. For example, if a person has a positive overall impression of a celebrity, they might attribute intelligence or kindness to that celebrity, even without specific evidence.

v) Journals

Ans) Journals, in an academic context, refer to periodical publications that contain scholarly articles, research findings, and reviews in various fields of study. Academic journals serve as a platform for researchers to share their work and findings with the academic community.

vi) Checklist

Ans) A checklist is a written or digital tool used to keep track of tasks, items, or steps in a systematic and organized manner. Checklists are commonly used in various fields, including aviation, healthcare, and project management, to ensure that important actions are not overlooked.

vii) Scattergraph

Ans) A scattergraph, also known as a scatter plot or scatter diagram, is a graphical representation of data points in a two-dimensional space. Each data point is represented by a dot, and scattergraphs are often used to visualize the relationship between two variables.

viii) Factor Analysis

Ans) Factor analysis is a statistical method used to identify and explore the underlying structure or patterns in a dataset. It is often employed in fields like psychology and social sciences to uncover the latent factors that influence observed variables.

ix) Contingency tables

Ans) Contingency tables, also known as cross-tabulation tables, are used in statistics to display the frequency distribution of two or more categorical variables. They help examine the relationships between variables and are commonly used in hypothesis testing and chi-squared analysis.

x) Degree of Freedom

Ans) In statistics, the degree of freedom (often abbreviated as "df") represents the number of values in the final calculation of a statistic that are free to vary. It is an important concept in hypothesis testing and is used to determine the appropriate statistical distribution to apply when analyzing data.

Q2) Differentiate between the following:

i) Random error & Systematic error

Ans) Comparison between Random error and Systematic error:

ii) Descriptive and Analytical Research Design

Ans) Comparison between Descriptive and Analytical Research Design:

iii) Stratified Random Sampling and Simple Random Sampling

Ans) Comparison between Stratified Random Sampling and Simple Random Sampling:

iv) Structured and Unstructured Questionnaire

Ans) Comparison between Structured and Unstructured Questionnaire:

v) Power test and Speed test

Ans) Comparison between Power test and Speed test: