Student Survey PDF

Shortly after the terrorist attacks in New York City and Washington, DC, in September of 2001, researcher Jennifer Lerner and her colleagues conducted an Internet-based survey of nearly 2,000 American teens and adults ranging in age from 13 to 88 (Ler...

Shortly after the terrorist attacks in New York City and Washington, DC, in September of 2001, researcher Jennifer Lerner and her colleagues conducted an Internet-based survey of nearly 2,000 American teens and adults ranging in age from 13 to 88 (Lerner, Gonzalez, Small, & Fischhoff, 2003). They asked participants about their reactions to the attacks and for their judgments of various terrorism- related and other risks. results were that the participants tended to overestimate most risks, that females did so more than males, and that there were no differences between teens and adults The study by Lerner and her colleagues is an example of survey research in psychology Constructing Surveys Issue: complexities involved in answering a seemingly simple survey question and highlights the cognitive processes and challenges respondents face when interpreting and responding to such a question 1. Interpreting the Question Ambiguity (Malabo) of Terms: Respondents must first determine the meaning of key terms like "alcoholic drinks" and "typical day." Alcoholic Drinks: Do these include all types of alcohol (beer, wine, liquor), or just hard liquor? Different interpretations can lead to different answers. Typical Day: Is the question referring to a regular weekday or a weekend? Drinking patterns often differ between weekdays and weekends, and this ambiguity could affect how the question is interpreted. 2. Retrieving Relevant Information from Memory What Information to Retrieve: Respondents must decide what memories to access to answer the question. Do they think about: Recent Occasions: Recalling specific recent instances of drinking. Detailed Recall: Trying to remember and count every drink from a specific timeframe (e.g., last week). General Beliefs: Relying on pre-existing beliefs or self-perceptions about their drinking habits (e.g., "I am a light drinker"). This step shows that the process of retrieving memories isn’t always straightforward. People might retrieve vague impressions or attempt to make calculations to estimate their behavior. 3. Formulating a Tentative Judgment Estimation Process: Once they retrieve relevant memories, respondents must estimate how many alcoholic drinks they consume in a "typical day." This estimation could be done by: Averaging: Dividing the number of drinks consumed over a recent week by seven to get an average per day.  General Impression: Giving a rough estimate (a guess that may not be very accurate) based on their memory and self-perception. 4. Interpreting the Response Options Understanding the Scale: The response options themselves introduce another layer of difficulty. What is "Average"?: Respondents may not have a clear sense of what constitutes "average" alcohol consumption. Without a reference point, it’s hard to assess where they stand relative to the "average." Relative Terms: Phrases like "somewhat more" or "a lot fewer" than average are vague and can be interpreted differently by different people. 5. Editing the Response (Social Desirability Bias) Self-Presentation: Respondents may decide to modify their answer to present themselves in a better light. For example, if someone believes they drink "a lot more than average," they may choose to report "somewhat more than average" to avoid judgment. Social Desirability Bias: This refers to the tendency to give socially acceptable answers, especially on sensitive topics like alcohol consumption. Cognitive Demands on Respondents: Cognitive Complexity: This entire process is cognitively demanding. Respondents must interpret, retrieve, judge, and report, all while potentially considering how they appear to the researcher. Risk of Inaccuracy: Each step introduces the possibility of error or bias. Misinterpretation of the question, difficulty recalling information, or reluctance to answer honestly can all lead to inaccurate survey responses. CONTEXT EFFECTS IN SURVEYS occur when factors other than the content of the survey questions influence how people respond. These effects arise from the context in which items are presented, including their order, wording, or the structure of response options 1. Item-Order Effects: The order in which survey questions are presented can impact how respondents interpret and answer subsequent questions. Example: Strack et al. (1988) conducted a study where they asked college students two questions: one about general life satisfaction and another about dating frequency. The results showed that: When life satisfaction was asked first, the correlation between life satisfaction and dating frequency was weak (−.12). When dating frequency was asked first, the correlation was much stronger (+.66). Explanation: When asked about dating first, students had that information more readily accessible in memory, and they were more likely to use it as a basis for judging their overall life satisfaction. This is an example of how context (the order of questions) can shape responses. 2. Response Options Effects: The range of response options provided can also influence how people answer. Example: When respondents were asked how often they were “really irritated,” they responded differently based on the response options available: If the options ranged from "less than once a year" to "more than once a month," respondents thought of major irritations and reported being irritated infrequently. If the options ranged from "less than once a day" to "several times a month," respondents thought of minor irritations and reported being irritated more frequently. Explanation: People interpret the response options differently depending on the scale provided, leading to variations in their answers. 3. Perception of 'Normal' or 'Typical' Responses: Respondents tend to view the middle response option as the most "normal" or "typical" and often select it if they consider themselves to be average. Example: People tend to report watching more TV if the middle response option is "4 hours" compared to when the middle option is "2 hours." Explanation: The middle option shapes perceptions of what is considered normal behavior, and respondents anchor their answers to this perceived norm. 4. Mitigating Context Effects: To reduce item-order effects, researchers can randomize or counterbalance the order in which questions are presented, particularly in online surveys where order can be easily manipulated. Example: A study showed that undecided voters were 2.5% more likely to vote for the first candidate listed on a ballot, simply because of the order. This highlights the importance of randomizing question or response order to prevent biased responses. Survey questions can be either open-ended or closed-ended Closed-Ended Questions These questions give specific answer options that participants must choose from. Examples: "How old are you?" Under 18 18 to 34 35 to 49 50 to 70 Over 70 "On a scale of 0 (no pain) to 10 (worst pain ever), how much pain are you in right now?"  "Have you ever been depressed for two weeks or more?" Yes / No Why use closed-ended questions? Useful when you already know the possible responses people might give. Easier to analyze because the answers can be quickly turned into numbers. Participants can answer them more quickly. Downsides of closed-ended questions: They are harder to write because you need to carefully think about the answer options. Closed-Ended Questions Types of Closed-Ended Questions For questions about categories (like gender or race), you list options for participants to choose from. For questions about opinions or behaviors, a rating scale is used. Rating scales are ordered sets of responses, like “Never, Rarely, Sometimes, Often, Always” or “Strongly Agree, Agree, Neutral, Disagree, Strongly Disagree.” 5-point scales are common for questions about frequency or single traits (e.g., how often something happens). 7-point scales are better for questions where there’s a range of feelings (e.g., liking something a lot vs. a little). Branching: For some questions, it helps to ask a general question first (e.g., "Do you like ice cream?") and then ask more detailed questions based on the response (e.g., "How much do you like ice cream?"). This makes responses more reliable. Key Tips: Use verbal labels (like “Strongly Agree” or “Always”) instead of numbers to make the options clear for participants. If helpful, you can use visual scales (like a line where participants mark their answer) to make it easier for them to show how strongly they feel. LIKERT SCALE a type of rating scale used to measure people's attitudes by having them indicate their level of agreement or disagreement with a series of statements Key Features of a Likert Scale Statements: A Likert scale presents a series of statements about a person, group, or idea. These statements may be either positive or negative, allowing for a balanced assessment of attitudes. Example: Suppose a researcher wants to measure attitudes toward online learning. They create the following statements: "Online learning is convenient." "I find it difficult to focus during online classes." "Online education is as effective as face-to- face education." Agreement Levels: Respondents indicate their level of agreement with each statement on a scale typically consisting of five points: Strongly Agree Agree Neither Agree nor Disagree Disagree Strongly Disagree Example: A student responding to the statement "Online learning is convenient" might select "Agree," which corresponds to a numerical value. Numerical Scoring: Each level of agreement is assigned a numerical value. For example: Strongly Agree = 5 Agree = 4 Neither Agree nor Disagree = 3 Disagree = 2 Strongly Disagree = 1 Example: If the student selects "Agree" for the statement "Online learning is convenient," it is given a score of 4. Reverse Coding: For negatively worded statements, the numerical scoring is reversed. This ensures that all responses are measured consistently. Example: For the statement "I find it difficult to focus during online classes" (a negative statement), a response of "Strongly Agree" would be coded as 1 (since agreeing with a negative statement reflects a less favorable attitude). Conversely, "Strongly Disagree" would be scored as 5. Summing the Scores: The total score across all items represents the overall attitude toward the subject being measured. Example: If a respondent scores 5 for "Online learning is convenient," 2 for "I find it difficult to focus during online classes" (reverse- coded to 4), and 3 for "Online education is as effective as face-to- face education," the total score would be 12 (out of a possible 15), indicating a generally positive attitude toward online learning. What Isn't a Likert Scale? A rating scale refers more generally to any numerical scale where people rate something, such as 0-to-10 scales for satisfaction or pain levels. These aren't Likert scales because they don't involve agreement with a series of statements. Example of a Rating Scale (Not Likert): "On a scale from 0 to 10, how satisfied are you with online learning?" This is simply a rating scale, not a Likert scale, because it's just one item asking for a rating, not multiple statements assessing agreement. In short, a Likert scale involves multiple statements and requires respondents to indicate their level of agreement. When reverse-coded items are used, it ensures all responses align in the same direction. writing effective questionnaire items using the BRUSO model to guide the process. BRUSO Model for Effective Questionnaire Items Brief Relevant Unambiguous Specific Objective 1. Brief Effective questionnaire items are concise and avoid long or overly complicated wording. This ensures that respondents can quickly understand and answer the questions. Example of a Poor Item: "To what extent do you believe that the current situation surrounding economic challenges and unemployment in your country is negatively affecting people's ability to maintain a stable quality of life?" Improved, Brief Version: "How much has unemployment affected quality of life?" Relevant The questions should be directly related to the research objectives. Avoid including unnecessary or intrusive questions, which can irritate respondents and lengthen the questionnaire. Example of a Poor Item: Asking about income or sexual orientation in a survey on exercise habits, which is irrelevant to the topic. Improved Version: Only ask about income or personal details if they directly relate to the research question (e.g., for studying socioeconomic impacts on exercise behavior). Unambiguous Ambiguity can lead to confusion, and different respondents may interpret the same question differently. Effective items are clear, with precise wording. Example of a Poor Item: "How often do you drink alcoholic beverages?" This question is ambiguous because "alcoholic beverages" and "often" are subjective. Respondents may have different interpretations of what counts as "alcohol" and "often." Improved Version: "How many alcoholic drinks (e.g., beer, wine, spirits) do you consume in a typical week?" Specific Questions should be focused on one concept at a time. "Double-barreled" questions ask about two different things in one item, which can confuse respondents and lead to unreliable data. Example of a Poor, Double-Barreled Item: "To what extent have you been feeling anxious and depressed?" This item combines two distinct emotions (anxiety and depression) into one question, but the respondent may experience one without the other. Improved Version: "To what extent have you been feeling anxious?" "To what extent have you been feeling depressed?" Objective Effective items are free from the researcher's opinions or biases. The question should not lead respondents toward a particular answer.  Example of a Poor, Leading Item: "Don't you agree that exercise is beneficial for health?" This phrasing implies that the "correct" answer is to agree. Improved, Objective Version: "To what extent do you believe that exercise is beneficial for health?" Additional Considerations: Pilot Testing: Conducting a pilot test allows you to see how people interpret your questions. This helps identify any potential misunderstandings before the questionnaire is widely distributed. Mutually Exclusive and Exhaustive Categories: Categorical questions should be mutually exclusive (no overlap) and exhaustive (cover all possible answers). Example of a Poor Item: "What is your religion? Christian Catholic Jewish Other" This is problematic because "Christian" and "Catholic" overlap. Additionally, there are many other religions not included here. Improved Version: "What is your religion? Protestant Catholic Jewish Other (please specify): ____________" Balanced Rating Scales: Rating scales should be balanced around a neutral midpoint, allowing for equally positive and negative responses. Example of an Unbalanced Scale:  "Unlikely | Somewhat Likely | Likely | Very Likely | Extremely Likely" This scale is skewed toward positive responses. Improved, Balanced Version: "Extremely Unlikely | Somewhat Unlikely | As Likely as Not | Somewhat Likely | Extremely Likely" Including or Omitting Neutral Options: Sometimes, researchers choose to omit a neutral or middle option to encourage deeper thinking. However, including a middle option on a bipolar scale (e.g., "Likely" vs. "Unlikely") allows for a more nuanced response if needed. Example Without a Neutral Option: "Agree | Disagree" Example With a Neutral Option: "Agree | Neither Agree nor Disagree | Disagree" Conclusion: Writing effective questionnaire items requires attention to detail and adherence to the BRUSO principles. By making items brief, relevant, unambiguous, specific, and objective, you maximize the reliability and validity of the data collected. Conducting a pilot test and carefully designing response categories are also essential steps in the process. Conducting Surveys Sampling in Psychological Research: Sampling is the process of selecting individuals from a population to study in research. In psychological research, it is vital to choose samples that can represent the broader population, allowing researchers to generalize their findings. Types of Sampling 1. Probability Sampling Probability sampling means the researcher can specify the likelihood that each person in the population will be selected for the sample. The goal is to make the sample representative of the population, enhancing the generalizability of the research. Examples of Probability Sampling: 1. Simple Random Sampling: Every individual in the population has an equal chance of being selected.  Example: A researcher creates a list of all students in a university and randomly selects 100 students by drawing names from a hat or using a computer. 2. Stratified Random Sampling: The population is divided into subgroups (strata) based on certain characteristics, and individuals are randomly selected from each subgroup. Example: If a researcher wants to ensure that their sample mirrors the racial makeup of the U.S., they would divide the population into racial categories (e.g., 12.6% African American, 5.6% Asian American) and randomly select individuals within each group. Types of Sampling 1. Probability Sampling Cluster Sampling: The population is divided into clusters (usually based on geography or institution), and clusters are randomly selected. Then, individuals within each selected cluster are randomly sampled. Example: A researcher studying small-town residents might randomly choose 10 small towns and then randomly select 20 individuals from each town. Types of Sampling Non-Probability Sampling Non-probability sampling occurs when the researcher cannot specify the probability that each member of the population will be selected. This type of sampling is more common in psychological research. Examples of Non-Probability Sampling: Convenience Sampling: The researcher studies individuals who are readily available and willing to participate. Example: A psychologist conducts a study on stress levels by surveying students from their own classes or colleagues at their workplace. Types of Sampling Non-Probability Sampling Non-probability sampling occurs when the researcher cannot specify the probability that each member of the population will be selected. This type of sampling is more common in psychological research. Examples of Non-Probability Sampling: Snowball Sampling: Current participants recruit additional participants. Example: In a study on coping strategies among cancer survivors, participants recommend other survivors they know to join the study. Types of Sampling Non-Probability Sampling Non-probability sampling occurs when the researcher cannot specify the probability that each member of the population will be selected. This type of sampling is more common in psychological research. Examples of Non-Probability Sampling: Snowball Sampling: Current participants recruit additional participants. Example: In a study on coping strategies among cancer survivors, participants recommend other survivors they know to join the study. Sampling bias occurs when the sample selected for a study is not representative of the entire population, leading to inaccurate results. Example: The 1936 Literary Digest Straw Poll The classic example of sampling bias comes from the Literary Digest poll conducted during the 1936 U.S. presidential election. The magazine incorrectly predicted that Alfred Landon would  defeat Franklin D. Roosevelt. The error occurred because the sample was drawn from telephone directories and lists of car owners, which overrepresented wealthier individuals—those more likely to support Landon. As a result, the poll was biased, not reflecting the voting preferences of the general population, many of whom were less affluent and favored Roosevelt. Non-Response Bias Non-response bias is a particular form of sampling bias that occurs when certain groups are less likely to respond to the survey, causing the results to be twisted on side. For example, in a study on alcohol consumption, if non-drinkers are less likely to participate, the results may suggest a higher rate of drinking in the population than is actually true. Example: Lahaut's Alcohol Consumption Study Researcher Vivienne Lahaut and her team found that only about half of their sample responded to a mail survey on alcohol consumption. By following up with non-responders, they discovered that many of them were abstainers (non-drinkers). If the study had only relied on the initial respondents, it would have inaccurately concluded that alcohol consumption was higher than it actually was. Methods to Address Non-Response Bias Researchers use several methods to minimize non-response bias: Increasing the response rate: Researchers can improve response rates by using strategies such as sending pre-notification messages, follow-up reminders, or incentives (e.g., small cash rewards) to encourage participation. CONDUCTING SURVEYS: METHODS AND RESPONSE RATES Researchers can conduct surveys in several ways, each of which has implications for response rates and potential biases. In-Person Interviews Highest response rates because of direct contact.  Important when the interviewer needs to observe the participant (e.g., mental health assessments). Costly and labor-intensive, limiting their use. Telephone Surveys Lower response rates than in-person interviews but still provide personal contact. Less expensive than in-person interviews. Use of telephone directories for sampling has become less effective due to the increasing use of cell phones over landlines. Mail Surveys Cost-effective, but they tend to have lower response rates, making them more vulnerable to non-response bias. Follow-up reminders and pre-notifications can help boost response rates. Internet Surveys Becoming more common due to their low cost and ease of construction. Response rates can vary depending on how respondents are recruited (e.g., email invitations vs. web postings). Can be difficult to obtain a random sample because not everyone uses the internet or the same websites. Studies show that internet-based findings are often consistent with traditional methods, although internet samples are not always fully representative of the general population. THE INTERVIEW METHOD IN PSYCHOLOGY Interviews involve a conversation with a purpose, but have some distinct features compared to ordinary conversation, such as being scheduled in advance, having an asymmetry in outcome goals between interviewer and interviewee, and often following a question-answer format.  Interviews are different from questionnaires as they involve social interaction. Unlike questionnaire methods, researchers need training in interviewing (which costs money). The interview method in psychology is a data collection technique where a researcher engages in direct conversation with individuals to gather information about their thoughts, experiences, and behaviors. It involves asking structured or open-ended questions to elicit responses that can provide insights into various psychological phenomena. Interviews can be used in clinical assessments, research studies, and therapeutic settings, allowing for in-depth exploration of topics and the subjective experiences of individuals. This method helps researchers understand subjective perspectives, obtain qualitative data, and gain a deeper understanding of human HOW DO INTERVIEWS WORK? HOW RESEARCHERS GATHER DATA FROM PARTICIPANTS THROUGH QUESTIONS  Researchers can ask different types of questions, generating different types of data. For example, closed questions provide people with a fixed set of responses, whereas open questions allow people to express what they think in their own words. Closed vs. Open Questions: Closed questions provide participants with a fixed set of responses (e.g., Yes/No or multiple-choice). These generate quantifiable, easy-to-analyze data, but they limit the depth of responses. Open questions allow participants to answer in their own words, offering richer, more detailed insights but require more effort to analyze. Interview Recording and Transcription: Researchers often record interviews to capture what was said. Later, the recording is transcribed into a written format (a transcript), which can be systematically analyzed. This allows for an accurate, detailed examination of the responses. Sensitive Topics and Method Choice: Interviews may not always be the best choice for exploring sensitive issues (e.g., truancy, discrimination). People might feel uncomfortable discussing such topics openly with an interviewer. In these cases, questionnaires, which can be completed privately, may be a better option as they offer more anonymity and comfort for the respondent.  Researchers can ask different types of questions, generating different types of data. 1. Types of Interviews: 1. Structured interviews: The researcher asks preset questions in a specific order. This is more rigid and ensures consistency across interviews, but it limits flexibility. 2. Unstructured interviews: These have a free-flowing, conversational style without a predetermined set of questions. This allows the interviewer to explore topics in-depth, but it can be hard to analyze due to variability in responses. 3. Semi-structured interviews: These are the most common in psychology research. The researcher follows a flexible interview guide, with some preset questions but allows room to explore topics that emerge during the conversation. This approach balances structure with the ability to capture detailed responses. Each method has its advantages and is chosen based on the research goals, the type of data needed, and the sensitivity of the topic being studied.  The interviewer will not deviate from the interview schedule (except to clarify the meaning of the question) or probe beyond the answers received. Replies are recorded on a questionnaire, and the order and wording of questions, and sometimes the range of alternative answers, is preset by the researcher. A STRUCTURED INTERVIEW IS ALSO KNOWN AS A FORMAL INTERVIEW (LIKE A JOB INTERVIEW). Strengths: Ease of Replication: Because structured interviews use the same set of closed questions for every participant, they can be easily repeated in future studies. This consistency allows researchers to compare results over time and with different groups. Quantifiability: The responses from structured interviews can be easily quantified (converted into numerical data). This makes it straightforward to perform statistical analyses, enhancing the reliability of the findings. Strengths: Reliability Testing: The fixed nature of the questions means researchers can test how reliable the results are. If different researchers conduct the same structured interview, they should yield similar results if the questions are clear and effective. Efficiency: Structured interviews can be conducted relatively quickly since the interviewer follows a set format. This efficiency allows researchers to conduct many interviews in a short period, increasing the likelihood of gathering a large sample size. Strengths: Representativeness: A larger sample size can lead to findings that are more representative of the broader population. This is important for generalizing the results, allowing researchers to make conclusions that apply to a wider group rather than just the participants in the study. LIMITATIONS: Lack of Flexibility: Structured interviews follow a strict interview schedule with predetermined questions. This means that interviewers cannot ask spontaneous or follow-up questions based on a participant's responses. The rigid format can restrict the depth of conversation and exploration of unexpected topics limited Detail in Responses: Since structured interviews primarily use closed questions (e.g., Yes/No, multiple-choice),  the responses tend to be brief and focused on specific answers. While this allows for easy quantification of data, it means that the information gathered lacks depth and detail. LIMITATIONS: Quantitative Focus: The data obtained is primarily quantitative, which may not capture the nuances of participants' thoughts, feelings, or motivations. As a result, researchers may miss out on understanding the reasons behind certain behaviors or opinions, limiting the richness of the insights Understanding Behavior: Because structured interviews do not explore qualitative aspects (such as why someone feels or behaves a certain way), researchers may struggle to fully comprehend the underlying reasons for a participant's responses. This can hinder a more comprehensive understanding of the topic being studied. In summary, while structured interviews offer consistency and ease of analysis, they can fall short in capturing the complexities of human behavior and experiences due to their lack of flexibility and reliance on closed questions.  Unstructured interviews are sometimes referred to as ‘discovery interviews’ and are more like a ‘guided conservation’ than a strictly structured interview. They are sometimes called informal interviews. Unstructured interviews are most useful in qualitative research to analyze attitudes and values. Though they rarely provide a valid basis for generalization, their main advantage is that they enable the researcher to probe social actors’ subjective points of view. THE COMPLEXITIES OF UNSTRUCTURED INTERVIEWS, PARTICULARLY THE ROLE OF INTERVIEWER SELF- DISCLOSURE AND ITS IMPACT ON THE INTERVIEW DYNAMICS Unstructured Interview Style: Unstructured interviews resemble informal conversations, allowing for a free-flowing exchange of ideas. Interviewers may choose to share personal experiences or thoughts (self- disclosure) to create a more relaxed atmosphere. Impact of Self-Disclosure: While self-disclosure can help build rapport and encourage openness from participants, it can also shift the focus away from the interviewee's perspective. The dynamic may change, making it harder for participants to provide their own accounts without being influenced by the interviewer’s views. Balancing Rapport and Neutrality: Striking a balance between being personable and maintaining neutrality is crucial. If the interviewer becomes too informal or shares too much, the interview may devolve into an ordinary conversation, leading to “consensus accounts,” where participants align their views with the interviewer’s rather than expressing their genuine thoughts. Potential Risks:  Excessive self-disclosure can be perceived as irrelevant or intrusive, especially if it touches on sensitive topics. Participants might feel uncomfortable, which could hinder the openness intended in the interview. Recommendations: it’s generally safer to avoid self-disclosure in interviews. If an informal style is employed, any disclosures should be made with careful judgment and experience. If participants ask for the interviewer’s opinions, the interviewer should clarify their role and defer those discussions to maintain the integrity of the interview. STRENGTHS OF UNSTRUCTURED INTERVIEWS Flexibility: Unstructured interviews allow the interviewer to adapt and modify questions based on the participant's responses. This means the conversation can flow naturally, enabling the interviewer to explore topics in greater depth as they arise, rather than sticking rigidly to a predetermined schedule. Qualitative Data: These interviews primarily generate qualitative data through open-ended questions. Participants can express their thoughts and feelings in their own words, leading to richer, more detailed responses. This depth helps researchers gain a deeper understanding of the participant's perspective on a situation. Increased Validity: Unstructured interviews enhance validity because they allow interviewers to probe deeper into responses. They can ask follow-up questions for clarification and encourage participants to elaborate on their answers. This interactive approach helps ensure that the data collected reflects the true beliefs and feelings of the interviewee,  rather than being constrained by fixed responses. participant-Driven Direction: The format allows the interviewee to guide the discussion to some extent, which can lead to the discovery of unexpected insights. This participant-driven approach ensures that the most relevant topics are explored, enhancing the richness of the data. Clarification Opportunities: Interviewers can clarify any confusion or misunderstandings during the conversation, which helps to ensure that the data collected is accurate and meaningful. This real-time interaction fosters a more open and engaging dialogue, contributing to a deeper understanding of the interviewee's views. Limitations OF UNSTRUCTURED INTERVIEWS Time-Consuming: Conducting unstructured interviews and analyzing the resulting qualitative data can be very time-consuming. Unlike quantitative data, which can be quickly summarized and analyzed, qualitative data requires thorough examination, often involving methods like thematic analysis. This process involves identifying patterns, themes, and insights within lengthy, detailed responses. Cost of Interviewers: Hiring and training skilled interviewers can be expensive compared to collecting data through questionnaires, which often require less personnel and training. Skilled interviewers need to possess certain abilities, such as establishing rapport with participants and knowing when to probe for deeper understanding, which can require considerable investment in training and resources Co-Construction of Data: Unstructured interviews inherently involve a collaborative process between the interviewer and the participant, meaning the data collected is influenced by both parties. Researchers’ agendas and the way questions are framed can shape the responses given. While open questions aim to minimize this bias, they cannot completely eliminate it. The interviewer's influence can lead to a focus on certain topics while sidelining others, potentially skewing the findings. Limited Remedies: Although open questions in unstructured interviews encourage detailed responses, they provide limited solutions to the issue of bias introduced by the interviewer. The subjective nature of the interaction can still affect the authenticity of the data collected.  In semi-structured interviews, the interviewer has more freedom to digress and probe beyond the answers. The interview guide contains a list of questions and topics that need to be covered during the conversation, usually in a particular order. Semi-structured interviews are most useful to address the ‘what’, ‘how’, and ‘why’ research questions. Both qualitative and quantitative analyses can be performed on data collected during semi-structured interviews. STRENGTHS OF SEMI- STRUCTURED INTERVIEWS: Respondent-Centered Flexibility: Semi-structured interviews create a more informal environment, allowing respondents to express their thoughts and feelings in their own words. Uniform Information with Depth: While they maintain a level of structure through predetermined questions, semi-structured interviews still provide  the flexibility needed to explore topics more deeply. This balance helps gather consistent data across interviews while allowing for individual insights. Exploration of Ideas: The flexible nature allows the interviewer to introduce new ideas or topics based on the respondent's answers. This adaptability can uncover unexpected themes and insights, enhancing the depth of the qualitative analysis. Probing for Clarity: Interviewers can ask follow-up questions to clarify or expand on responses, which helps to ensure that the data collected is reliable and comprehensive. This probing ability allows for a deeper understanding of the respondent's perspectives. LIMITATIONS OF SEMI-STRUCTURED INTERVIEWS 1. Interviewer Skill Dependency: The quality of the data relies heavily on the interviewer’s skills. Effective probing and maintaining neutrality are crucial to avoid biasing responses. In structured interviews, the format provides more consistency, whereas semi-structured formats depend on the interviewer’s ability to navigate discussions skillfully. Example of Interviewer Skill Dependency in Semi-Structured Interviews: Imagine a researcher conducting semi-structured interviews to explore students' mental health during online learning. If the interviewer is skilled, they might ask open-ended questions and effectively probe for deeper insights, allowing students to share their experiences in detail. However, if the interviewer lacks experience or becomes too directive, they might unintentionally bias responses. For instance, if they express surprise or disbelief at a student’s comment about feeling isolated, the student may feel discouraged from elaborating on that experience. This could lead to a more superficial understanding of the issue.