Research method. c17 2nd sem IISWBM.pptx

Hey there! I just wanted to share this awesome PowerPoint presentation with you. It has all the answers to the theoretical questions for the second semester C17 paper at IISWBM, Calcutta University. This material was taught by Sumanti Mam and JD Sir, and it really helped our whole batch ace the theory questions in the 2nd semester of 2024. I hope it will be just as helpful for the new junior students too.
By the way, I'm an MBA day student for the 2023-2025 academic year.
Connect with me: www.linkedin.com/in/sagnik-sanyal

1. Research methodology c17 solutions of theory question Prepared By SAGNIK SANYAL MBA(DAY) IISWBM BATCH 23-25

2. why ratio scale considered as most powerful scale? 1. Absolute Zero Point Definition: The ratio scale has a true zero point that indicates the absence of the quantity being measured. This means that zero on this scale is meaningful and represents a complete lack of the variable. Example: Weight is measured on a ratio scale because zero weight means no weight at all. 2. All Mathematical Operations Capability: The ratio scale supports all mathematical operations, including addition, subtraction, multiplication, and division. Example: You can say that 20 kg is twice as heavy as 10 kg, or that 15 kg is 5 kg more than 10 kg. These operations are meaningful and accurate. 3. Comparison of Magnitudes Definition: Because the ratio scale has a true zero and equal intervals, it allows for comparisons of absolute magnitudes and ratios. Example: In measuring time, you can compare two time periods and say that 30 minutes is twice as long as 15 minutes. 4. Highest Level of Precision Accuracy: Ratio scales provide the highest level of precision and accuracy in measurement, making them ideal for scientific and engineering applications. Example: Measuring distances in meters or lengths in centimeters allows for highly precise calculations and comparisons.

3. Types of Data Collected through Observation Method Behavioral Data Definition: Behavioral data refers to observable actions, reactions, and interactions of subjects within their environment. Examples: Shopping Behavior: Observing how customers navigate through a store, which products they pick up, and their purchasing decisions. Workplace Interaction: Watching how employees interact with each other and use office equipment. Classroom Engagement: Noting students' participation, attention levels, and interactions during a lesson. Non-Behavioral Data Definition: Non-behavioral data refers to information that does not involve observing actions but instead focuses on the physical and situational aspects of the environment or conditions. Examples: Physical Environment: Recording the layout of a store, office, or classroom. Condition of Objects: Observing the wear and tear of machinery or equipment. Process Flow: Documenting the sequence of operations in a manufacturing line or service delivery.

4. Differences between Staple Scale and Semantic Differential Scale : Response Options : The Semantic Differential Scale presents pairs of opposite adjectives or phrases with a rating scale between them, allowing participants to indicate their position on a continuum. In contrast, the Staple Scale presents pairs of opposite adjectives or phrases without a numerical scale, requiring participants to choose one option from each pair without indicating intensity. Interpretation : While both scales require interpretation, the Semantic Differential Scale focuses on understanding the meaning associated with each endpoint, enabling a nuanced understanding of attitudes or perceptions. In contrast, the Staple Scale provides direct comparisons between the options in each pair, facilitating quick assessments without the need for interpretation. Granularity : The Semantic Differential Scale offers more granularity in responses by providing a range of options along a continuum, allowing for subtle distinctions in attitudes or opinions. On the other hand, the Staple Scale offers limited granularity with fixed choices, making it suitable for rapid assessments but less suitable for capturing nuanced differences. Scalability : Semantic Differential Scale can be scaled according to the number of items or dimensions being measured, allowing for flexibility in research design. In contrast, the Staple Scale typically involves a fixed set of pairs, limiting its scalability and adaptability to different contexts. Example : An example of a Semantic Differential Scale could be assessing perceptions of a brand using pairs like "innovative" vs. "conventional". In contrast, a Staple Scale might assess product preferences using pairs like "modern design" vs. "traditional design", where participants choose one option from each pair.

5. What are the types of research designs discuss about them?

6. D efine reliability and validity of measuring instrument Reliability of a Measuring Instrument: Consistency: Reliability refers to the consistency or stability of measurements obtained from a measuring instrument. A reliable instrument should yield consistent results when administered repeatedly under similar conditions. Internal Consistency : One aspect of reliability is internal consistency, which assesses the extent to which different items within the instrument measure the same underlying construct. High internal consistency indicates that items are measuring the same thing. Test-Retest Reliability: Another aspect of reliability is test-retest reliability, which assesses the stability of measurements over time. A reliable instrument should produce similar results when administered to the same individuals on two different occasions, assuming no real change in the construct being measured. Inter-Rater Reliability : For instruments involving subjective judgments or observations, inter-rater reliability is important. It measures the degree of agreement among different raters or observers when using the instrument. A reliable instrument should produce consistent results regardless of who administers it. Validity of a Measuring Instrument: Accuracy: Validity refers to the extent to which a measuring instrument accurately measures the construct it is intended to measure. A valid instrument should provide results that are true representations of the construct of interest. Content Validity: Content validity assesses the extent to which the items or questions in the instrument adequately represent the entire domain of the construct being measured. It ensures that the instrument covers all relevant aspects of the construct. Criterion Validity : Criterion validity examines the extent to which scores on the instrument correlate with scores on a criterion measure that is already established as valid. It demonstrates the instrument's ability to predict or correlate with an external criterion. Construct Validity: Construct validity assesses the extent to which the instrument measures the theoretical construct it claims to measure. It involves examining the relationship between scores on the instrument and other measures or behaviors that are theoretically related to the construct.

7. &

8. Each sampling method has its own advantages and limitations, and the choice of method depends on factors such as the population characteristics, research objectives, and available resources. Simple random sampling ensures each member of the population has an equal chance of being selected, while systematic sampling is efficient and less prone to human bias. Stratified sampling ensures representation from all subgroups, while cluster sampling is efficient for large populations with natural groupings. Multistage sampling combines multiple sampling methods to achieve representativeness and efficiency in complex sampling scenarios.

9. Nonprobability sampling involves selecting participants for a study in a non-random manner, often based on convenience or judgment. It is important in research methodology because it allows researchers to explore phenomena where random sampling is impractical or impossible, offering insights into specific populations or contexts with limited resources.

10. A s management trainee conduct survey to measure the consumption and preference for different brand of coffee in I ndian households using nominal,ordinal,interval, ratio scale? Nominal Scale: a. Which brand of coffee do you usually purchase? (Options: A, B, C, D) b. Which brand of coffee do you prefer the most? (Options: X, Y, Z, Other) Ordinal Scale: a. Please rank the following coffee brands based on your consumption frequency: (Options: A, B, C, D) b. On a scale of 1 to 5, where 1 is "strongly dislike" and 5 is "strongly like," how would you rate your preference for each coffee brand? (Options: X, Y, Z, Other) Interval Scale: a. On a scale of 1 to 10, where 1 is "never" and 10 is "daily," how often do you consume Brand A coffee? b. Please rate your satisfaction with the taste of Brand B coffee on a scale of 1 to 7, where 1 is "very dissatisfied" and 7 is "very satisfied." Ratio Scale: a. How many cups of Brand C coffee do you consume on an average day? b. How much money do you spend on Brand D coffee per month on average? (Options: Rs. 0-500, Rs. 501-1000, Rs. 1001-2000, Rs. 2001 and above)

11. A s management trainee conduct survey to measure the consumption and preference for different brand of cookies in I ndian households using constant sum , comparative sum, staple scale. Constant Sum : a. Allocate 100 points among the following cookie brands based on your consumption preference: (Options: Brand A, Brand B, Brand C, Brand D) b. Distribute 50 points among the cookie brands to indicate your preference for each: (Options: Brand X, Brand Y, Brand Z, Brand W) Comparative Sum : a. Compare the taste of Brand A cookies to Brand B cookies and allocate 10 points accordingly: (Options: Brand A better, Brand B better, Equal) b. Compare the packaging of Brand X cookies to Brand Y cookies and distribute 20 points to indicate preference: (Options: Brand X better, Brand Y better, Equal) Staple Scale : a. Rate the sweetness level of Brand C cookies on a scale of 1 to 5, where 1 is "too sweet" and 5 is "not sweet enough." b. Rate the crunchiness of Brand Z cookies on a scale of 1 to 7, where 1 is "not crunchy" and 7 is "extremely crunchy."

12. PROJECTIVE TECHNIQUE Projective techniques are qualitative research methods used to uncover underlying thoughts, feelings, and motivations of individuals by having them respond to ambiguous stimuli. Participants project their own perceptions, attitudes, and experiences onto the stimuli, revealing subconscious or latent aspects of their personality or mindset. Characteristics include ambiguity in stimuli, allowing for diverse interpretations; indirect questioning, bypassing conscious defenses; and rich, qualitative data, offering deep insights into participants' inner worlds. Projective techniques foster creativity and spontaneity, enabling researchers to explore complex psychological constructs and understand subconscious processes in individuals' responses. An extraneous variable is any variable other than the independent variable in a study that can affect the dependent variable. These variables are not of primary interest but can confound the results if not controlled. Researchers aim to identify and minimize the influence of extraneous variables to ensure that changes in the dependent variable are due to the manipulation of the independent variable. Ex-post facto research , also known as causal-comparative research, is a type of non-experimental research design in which the researcher observes the effects of naturally occurring differences in independent variables on the dependent variable. Unlike experimental research, where the independent variable is manipulated by the researcher, ex-post facto research observes and compares existing differences without manipulation. It aims to establish causal relationships by analyzing the effects of variables that the researcher cannot manipulate due to ethical or practical constraints.

13. Discuss advantage and disadvantage of interviewing as data collection method?

14. Discuss the difference Administrative questions, Classification questions, Target questions ( structured or unstructured)

15. What is difference between Exploratory, Descriptive , Diagnostic, Experimental research?

16. Explain the types of Experimental Design Before and After Without Control Design: Participants are measured on a dependent variable before and after exposure to the experimental treatment. There is no control group or comparison condition. Example: Measuring students' test scores before and after implementing a new teaching method without a control group. After-Only with Control Design: Participants are randomly assigned to either an experimental group (receiving the treatment) or a control group (not receiving the treatment). Both groups are measured on the dependent variable after the treatment is administered. Example: Randomly assigning participants to either receive a new medication (experimental group) or a placebo (control group), and then measuring their symptom severity after treatment. Before and After with Control Design: Participants are randomly assigned to either an experimental group or a control group. Both groups are measured on the dependent variable before and after the treatment is administered. Example: Randomly assigning students to receive extra tutoring (experimental group) or regular instruction (control group), and then measuring their test scores before and after the tutoring program. Completely Randomized Design: Participants are randomly assigned to different experimental conditions or treatment groups. Each participant has an equal chance of being assigned to any experimental condition. Example: Randomly assigning individuals to receive either a new therapy, a placebo, or no treatment for anxiety . Randomized Block Design: Participants are grouped into blocks based on certain characteristics (e.g., age, gender). Within each block, participants are randomly assigned to different experimental conditions. Example: Grouping participants based on their level of anxiety (low, moderate, high) and then randomly assigning them to receive different anxiety treatments. Latin Square Design: Involves a square grid where each row and column represents a different treatment condition. Each treatment appears once in each row and column, ensuring that each treatment is tested in combination with every other treatment an equal number of times. Example: Testing the effectiveness of different teaching methods (A, B, C, D) across different classrooms (1, 2, 3, 4) using a Latin square design to control for classroom effects. Factorial Design: Involves manipulating two or more independent variables simultaneously to examine their effects on the dependent variable. Allows researchers to study the main effects of each independent variable as well as any interaction effects between them. Example: Investigating the effects of both medication dosage (low vs. high) and therapy type (cognitive-behavioral vs. psychodynamic) on depression symptoms .

17. Explain how log 10 transformation is applied by the researchers to transform highly positive skewed distribution data and highly negative skewed data into near normal data. Logarithmic transformations, such as the log10 transformation, are commonly used by researchers to address issues of skewed distributions and non-constant variance in data. Here's how researchers apply log10 transformation to transform highly positively skewed and highly negatively skewed data into near-normal data, along with examples: Highly Positively Skewed Data: Issue: Highly positively skewed data typically have a long tail to the right, with most values clustered at the lower end of the distribution. Transformation : Apply a log10 transformation to compress the range of values, reducing the influence of extreme values and bringing the distribution closer to normality. Example: Consider a dataset of income levels where most individuals earn relatively low incomes, but a few individuals earn extremely high incomes. Applying a log10 transformation to the income data would compress the range of incomes, making the distribution more symmetrical and normally distributed. Highly Negatively Skewed Data: Issue : Highly negatively skewed data have a long tail to the left, with most values clustered at the higher end of the distribution. Transformation : Apply a log10 transformation to expand the range of values, reducing the influence of extreme low values and bringing the distribution closer to normality. Example: Suppose we have a dataset of reaction times in milliseconds, where most observations are clustered around fast reaction times but a few observations have extremely slow reaction times. Applying a log10 transformation to the reaction time data would expand the range of values, making the distribution more symmetrical and normally distributed.

18. How number of categorical variables and level of categorical variables determine the choice of t test and Anova while performing hypothesis testing? Number of Categorical Independent Variables: T-test: Used when there is only one categorical independent variable with two levels . This allows you to compare the means of two groups defined by the independent variable. ANOVA: Used when there are one or more categorical independent variables, but with more than two levels in at least one variable. ANOVA can handle comparisons of means across multiple groups defined by these variables. Focus of the Hypothesis: T-test: Focuses on comparing the means of two groups defined by the categorical independent variable. It's ideal for simple comparisons. ANOVA: Can analyze the effects of a single categorical independent variable with multiple levels ( one-way ANOVA ) or the interaction effects of two or more categorical independent variables ( two-way ANOVA or higher). It allows for a more comprehensive analysis of how the categorical variables influence the dependent variable. Example: You want to compare the average height of males and females (one categorical variable with two levels). You would use a t-test . You want to compare the average exam scores of students across three different teaching methods (one categorical variable with three levels). You would use a one-way ANOVA . You want to see if the effect of teaching method on exam scores depends on the student's gender (two categorical variables). You would use a two-way ANOVA .

19. Restaurant owner is interested to know how the customers feel about the service they receive at his restaurant. construct a semantic differential scale with 8 bipolar adjectives for that . mention the process the of analysis of the data collected from 50 customers. Friendly - Unfriendly This measures how welcoming and approachable the staff were. Attentive - Inattentive This measures how well the staff paid attention to your needs. Knowledgeable - Unknowledgeable This measures how knowledgeable the staff were about the menu and restaurant. Prompt - Slow This measures how quickly you were served and how long you had to wait. Efficient - Inefficient This measures how smoothly and organized the service was. Helpful - Unhelpful This measures how willing and able the staff were to assist you. Professional - Unprofessional This measures how courteous and well-trained the staff appeared. Accommodating - Unaccommodating This measures how flexible and willing the staff were to address your requests. Process Data Cleaning: Check for missing values, inconsistencies, and outliers. Coding: Assign numerical values to the Likert scale responses (1-7).Descriptive Statistics: Calculate average scores for each adjective pair. Visualization: Create bar charts or heatmaps to see how customers rated each service aspect. Optional: If applicable, use statistical tests to compare ratings across different demographics (e.g., age groups).

20. Identify the relevant population for following researchers and suggest the appropriate sampling design to investigate the issues explaining why they are appropriate: Q1. The general manager of firm wants to investigate the relationship between drug abuse and dysfunctional behavior of blue collar worker in a particular plant. Relevant Population: Blue-collar workers employed at the specific FR plant the general manager oversees. Appropriate Sampling Design: Stratified Random Sampling Explanation: Generalizability of the findings is crucial here. A random sample from the entire FR company's blue-collar workforce wouldn't be appropriate because it wouldn't reflect the specific plant's workers. Stratified random sampling involves dividing the population (all blue-collar workers in the plant) into subgroups (strata) based on relevant characteristics, like departments or shifts. Then, a random sample is drawn from each subgroup. Benefits of Stratified Sampling: Ensures representation from different subgroups within the plant's workforce. Provides more accurate data about the relationship between drug abuse and dysfunctional behavior specific to that plant's workers. Q2. An administrator wants to assess the reactions of employees to a new and improved health benefits scheme that requires a modest increase in the premiums to be paid by the employees for their families. Relevant Population: All employees of the organization who are eligible for the health benefits scheme. Appropriate Sampling Design: Simple Random Sampling or Stratified Random Sampling (depending on the scenario) Explanation: Simple Random Sampling: This is the most straightforward approach if the employee population is homogenous (similar demographics and benefit needs). Each employee has an equal chance of being selected, ensuring unbiased representation. Stratified Random Sampling: This might be preferable if there are significant subgroups within the employee population with potentially different reactions to the scheme. Strata could be based on factors like department, seniority, or family size. A random sample is then drawn from each subgroup. Choosing Between Simple and Stratified Sampling: If the increased premium primarily affects specific employee groups (e.g., larger families), stratified sampling ensures their voices are heard. If the impact is likely similar across the workforce, simple random sampling is efficient. Benefits of Random Sampling: Provides a representative sample of employee sentiment towards the new scheme. Allows for generalization of findings to the entire population of eligible employees .

21. Relationship Description Example Advantages Disadvantages Complete Observer (Unobtrusive) The researcher observes the participants without interacting or influencing their behavior. Participants are unaware they are being studied. Observing birds in their natural habitat using a hidden camera. - Minimizes observer bias. - Captures natural behavior. - Limited data on motivations and context. - Ethical concerns about deception. Passive Observer The researcher observes the participants without interacting, but their presence is known. Participants may alter their behavior slightly. Studying children in a classroom setting while sitting quietly in the back. - Less disruptive than complete observer. - Allows for recording detailed observations. - Potential for observer bias due to participant awareness. - Limited data on internal states. Active Observer The researcher observes and interacts with participants to a limited extent, asking clarifying questions or taking notes. Participating in a fitness class to observe workout routines and group dynamics. - Gain deeper insights into participant experiences. - Allows for clarification of observed behaviors. - Increased risk of observer bias due to interaction. - May alter participant behavior more than passive observer. Complete Participant (Covert) The researcher fully participates in the setting, concealing their research identity. Undercover police officer investigating drug activity in a gang. - Gaining in-depth understanding of the participant world. - Ethical concerns about deception and potential harm. - Legal issues depending on the research context. Open Participant The researcher openly discloses their research role and interacts with participants as a fellow member of the group. Joining a book club to study reading habits and group discussions. - Builds trust and rapport with participants. - Allows for open-ended questioning and clarification. - May influence participant behavior due to researcher presence. - Limited access to private or sensitive interactions. Discuss different types of observer participant relationship in observational studies?

22. how does org specify the research topic from huge area of research concern ?

23. Indicate the types of measurement scale and appropriate questions of the following: 1. customer preferences of refrigerator color. 2. importance of after sale services to customer. 3. budget of buying refrigerator. 1. Customer Preferences of Refrigerator Colour Type of Measurement Scale : Nominal Scale Appropriate Questions : Nominal : "Which colour do you prefer for your refrigerator? (Choose one)" Options: Red, White, Black, Silver, Blue Nominal : "Select your top 3 preferred refrigerator colours from the list below." Options: Red, White, Black, Silver, Blue 2. Importance of After Sale Services to Customer Type of Measurement Scale : Ordinal Scale Appropriate Questions : Ordinal : "How important is after-sale service to you when buying a refrigerator? (Rank from 1 to 5)" Options: 1 (Not Important), 2 (Slightly Important), 3 (Moderately Important), 4 (Very Important), 5 (Extremely Important) Ordinal : "Please rank the following aspects of after-sale service in order of importance (1 being the most important and 5 being the least important)." 3. Budget for Buying a Refrigerator Type of Measurement Scale : Ratio Scale Appropriate Questions : Ratio : "What is your budget for buying a refrigerator? (Please specify the amount in your local currency)" Answer: (Open-ended numerical response) Ratio : "Indicate the price range you are willing to spend on a new refrigerator." Options: Less than $300, $300 - $500, $500 - $700, $700 - $1000, More than $1000

24. Research Design To test these hypotheses, researchers could conduct a study involving cancer patients who are categorized based on the stage at which they are diagnosed (early vs. advanced) and the level of care provided by nurses (high vs. low). The recovery outcomes of these patients would be analyzed to determine the impact of early diagnosis and nursing care on recovery success. Variables Independent Variables : Early and correct diagnosis by the doctor. Nurse's careful follow-up of the doctor's instructions. Dependent Variable : Successful recovery of cancer patients under treatment. Hypotheses Development To examine the impact of the identified variables on the successful recovery of cancer patients, we can develop the following hypotheses: Hypothesis 1: Null Hypothesis (H0) : Early and correct diagnosis by the doctor does not have a significant effect on the successful recovery of cancer patients. Alternative Hypothesis (H1) : Early and correct diagnosis by the doctor has a significant effect on the successful recovery of cancer patients. Hypothesis 2: Null Hypothesis (H0) : Nurse's careful follow-up of the doctor's instructions does not have a significant effect on the successful recovery of cancer patients. Alternative Hypothesis (H1) : Nurse's careful follow-up of the doctor's instructions has a significant effect on the successful recovery of cancer patients.