idk

Lakukan tugas rumah & ujian kamu dengan baik sekarang menggunakan Quizwiz!

Response bias sources

1. Deliberate falsification 2. Unconscious misrepresentation 3. Inability to respond accurately

Testing for reliability

1. Differences in split samples a. Ask 200 people the same question. Split into 2 groups compare averages. 1. Predictivness (alternate forms) relates to the infinite number of questions appropriate for assessing any underlying construct. 2. Consistency with other answers 3. Consistency among coders

limitations of profile analysis

1. Only a few brands can be depicted 2. Attributes may not be independent 3. Profiles may not be weighted by attribute importance

Main sources of survey errors

1. Random sampling error 2. Systematic error

A measure is effective if it is

1. Reliable—or free from random error. Addresses the random and systematic error component of measurement. 2. Valid—or measures what is intended. More theoretical notion. They want to determine whether the measure is measure ring what its supposed to measure. Addresses the accurate- indicator dimension of differences within measures. 3. Generalizable—or extendable to the population 4. Sensitive—or calibrates individual differences

A measure is effective if it is

1. Reliable—or free from random error. Addresses the random and systematic error component of measurement. 2. Valid—or measures what is intended. More theoretical notion. They want to determine whether the measure is measure ring what its supposed to measure. Addresses the accurate- indicator dimension of differences within measures.. what its supposed to measure 3. Generalizable—or extendable to the population 4. Sensitive—or calibrates individual differences

Respondent error components

1. Response bias 2. Non-response error

Semantic Differential Scales

A Semantic Differential scale is a scale that contains a series of bipolar rating items, where bipolar adjectives anchor the endpoints for each item. extremities, not complete opposites

Q10. The question "I love watching the ads during the Super Bowl: 1) Strongly disagree, 2) Disagree, 3) Neither agree nor disagree, 4) Agree, 5) Strongly agree" is an example of a <-blank-> question. Which response below best fills in the blank.

A) Dichotomous. B) Likert scale. C) Open-ended. D) Multiple-choice. E) Semantic differential.

Q12. In questionnaire, a question such as, "Do you like Wendy's hamburgers and fries?" suffers from the drawback of A) being a leading question. B) being unclear and imprecise. C) Asking two questions at once. D) A,BandCaretrue. E) A and C are both true. F) A and B are both true. G) B and C are both true. Answer C)—Here two questions—one on hamburgers the other on fries—are being asked. See page 163 of the text, slide 15 of the chapter 9 slides, and slides 17 and 31 of the session 9 class slides.

C

"When will the technology exist to allow humans to live on Mars?"

C) is complex and something respondents may be unfamiliar with.

Reliability

• Degree to which measures are free from random error and therefore yield consistent results • Consistent, dependable reliable, not valid all close to the left vald not realible every where even

Q11. Non-response error A) Occurs when the survey answers from those who take the survey are influenced by those who refuse to take the survey. B) Can be reduced by increasing the response rates. C) Occurs when the responses of survey takers would have differed significantly if those who refused to take the survey had taken the survey. D) A, B, and C are correct E) Both B and A are correct. F) Both B and C are correct. G) Both A and C are correct.

F)—Non-response error occurs due to nonparticipation of some eligible respondents in a study, and when there would be a significant difference between the responses of those who responded to the study and those who don't. By reducing non-participation, or what is the same thing—increasing the response rate, non-response can be reduced, if not entirely eliminated. See pages 112 and 115 of the text, and slides 17 and 21 of the chapter 7 slides, and slides 23 and 30 of the session 7 class slides.

A B C

Feelings doing learnign

What is the first step in structuring a questionnaire?

First step towards ensuring that good questions are asked is to plan what is to be measured.

limitations of SD scales

Limitations of SD Scales 1. Respondents tend to misuse them 2. They're difficult to construct a. Limited to few words. 3. Negation doesn't necessarily mean the opposite a. Black and white opposties, or black and then no colors

Telephone surveys

Many potential respondents have encountered unscrupulous telemarketers pretending to conduct consumer research as part of a sales pitch. As a result, telephone interviews have dropped a lot in the last 20-30 years. 1. Call back potential respondents who aren't answering on different days at different times 2. Substitute other respondents for respondents who aren't home or wont answer 3. Called id telephone response rate very low

Minimizing Random Sampling Error

Minimizing Random Sampling Error • Should be minimal if you draw a sufficiently large sample o If you draw an insufficiently large sample, its possible that your particular sample will be idiosyncratic in some way. • Focus should be more on systematic error o Random sampling error is uncontrollable.

What is response rate?

Potential sample= completed responses + non responses

response bias

Response Bias • is bias that occurs when respondents tend to answer questions with a certain slant that consciously or unconsciously misrepresents the truth

sensitivity

Sensitivity determines whether a measure accurately detects differences among people. Sensitivity is critical for researchers trying to classify people into different groups, as would be the case for segmentation analysis,

Self-Administered, Paper-and-Pencil Surveys

The ever-popular self-administered questionnaire- the least personal approach-can be delivered by conventional mail service or by evermore popular electronic means (via web browser, email or interactive kiosk)

Hierarchy-of-Effects Model

The hierarchy of effects model is a representation of how communication works and is used by advertisers to make effective advertising. Most enduring L F D awareness 1.aided 2.unaided knowledge (belief) 1. attribute assessment 2. perceptions Affect 1`.like./disklike 2. first choice 3. consideration set 4. intentions 5. behavior

Generalizability

Whether the results of a study can be extended to a broader population or set of circumstances. Stogner the evidence to a larger population This is achieved over time through multiple studies.

measurement instrument bias

error from the design of questionnaire

random sampling error

error that results from chance variation

systematic error

error that results from problems or flaws in the execution of the research design

input error

errors that happen when data is being transferred from survey sheets to a computer

interviewer error

interviewer influences the respondent

sample design error

systematic error that results from an error in the sample design or sampling procedures

Misrepresentation unconscious

1. Acquiescence bias a. Or yeah saying is a type of response bias due to some people's tendency to agree with all questions or to concur with a particular position. Averages will be exagereated 2. Extremity bias a. "strongly disagree" or "strongly agree" 3. Auspices bias a. Also called sponsor bias is caused by respondents being influenced by the organization conducting the study. 4. Interviewer bias a. As it sounds because presence of an interviewer. Embarrassed. Have you done drugs 5. Subtle source of cues a. Skilled interviews avoid giving such cues. 6. Appearance of incompetence 7. Insufficient or poor probing a. Can you say more or ask another question giving answers unconciously

Two other pencil and paper surveys

1. Administered surveys a. Used to survey organization employees (sales persons of a sales force) and require managerial assistance for their implementation. (Administered surveys differ in the way that questionnaires are distributed to respondents. b. A researcher uses a company's managers to distribute and subsequently collect questionnaires from their employees. In essence, researches use managers as highly persuasive solicitors. Because respondents receive the same questions in the same order, their answers can be aggregated and meaningful subgroup comparisons can be made either within or between time periods. Administered surveys are popular for studying employees within a single major corporation; for example, to assess their attitudes about customers. Also, though useful for studying a large sales force, administered surveys rarely are used in consumer research. 2. Publication inserts and fax surveys a. Ineffective, outmoded, frequently resulting in biased samples. NOT recommended b. Publication inserts are ineffective because they are expensive and often yield biased samples. However, they can e useful for preliminary studies of hard-to-identify-and-access populations, such as spelunkers who subscribe to some magazine. c. Fax technology is old, relatively expensive, and awkward; it has been superseded by other electronic delivery technologies. Both insert questionaries' and unsolicited faxes may be mistaken for junk material or may go unnoticed.

Disadvantages of Open-Ended Questions

1. Articulate respondent bias 1. Articulate respondent bias 2. Interviewer bias a. Interviewers aren't alike. Thus interview differences can introuduce additional response bias. 3. Hard-to-record answers a. Have to scribble it down. 4. Coding inconsistency and difficulty a. Takes a lot of time and effort. Examine every response to every question, develop b. This disadvantage more than any other, is the one that causes us to urge you to use close-ended question if possible. They may take 5. Reduced cross-study comparability a. Open ended questions are more difficult to use for cross-study comparisons because choices and contexts change over time. 6. Complexity and costliness a. Open ended question are more costly because their best use requires live - and expensive- interviews.

Comparing Open-Ended and Close-Ended Questions

1. BIASED questions defeat the purpose of MR 2. Good questions are unbiased. 3. They should measure respondents' attitudes, preferences, and behaviors accurately 4. First step towards ensuring that good questions are asked is to plan what is to be measured. 5. After you identify and understand your targeted respondent youll have a better sense for how to write your questions.

Encouraging respondent cooperation

1. Be supportive or helpful a. Helpful when the interviewer is professional and plite and wants to be supportive of basic university research. 2. Emphasize social interaction a. Think about old people 3. Remunerate respondents a. Dollar or free pencil 4. Satisfy curiosity

Boosting Response Rates to Minimize Nonresponse Error For Telephone Surveys

1. Call Back a. Different days b. Different times 2. Substitute a. Respondent not home or not available

Disadvantages of close-ended questions

1. Can't obtain in-depth responses 2. Poor at providing new insights 3. More difficult to write because must anticipate all possible anwsers 4. Answers may not fully reflect respondents' attitudes a. Could be spot on, vague, or something in between 5. Categories hint at correct answers a. Your suggesting correct answers.

Advantages of close-ended questions

1. Communication skills of respondents are less critical a. Wont struggle to answer them because you can only select 2. Speedy response 3. Easier to answer 4. Responses can be quickly coded, entered, & analyzed 5. Interviewing skills are less important a. Either little or no interviewing skill is needed to administer close-edned questions, which such questions dominate self-administered surveys

Testing for validity

1. Content or face validity a. This type of validity merely assesses whether your questionnaire items adequately represent the underlying construct of interest. 2. Predictive validity a. Also knows as criterion validity, relates to where a measure is predictive of something 3. Convergent validity a. If you have a strong sense that you are measuring the correct thing, you should be able to develop alternative measures, and the scores on those measure should be consistent. 4. Discriminant validity 5. Nomoligical validity a. Measure to be related in a proper theoretical manner to other relevant marketing constructs.

mail surveys

1. Create an effective cover letter for your survey 2. Ensure that your questions are interesting and not overly taxing to answer. 3. Send follow ups 4. Reveal respected sponsor incentives

Planning What is Measured

1. Decide what needs to be learnd 2. Who should be questioned 3. How the data should be collected 4. How questions should be structured

reasons for using liker scales 1.

1. Easy to create 2. Familiar to respondents 3. Easy for respondents to answer 4. versatile-can use variety of descriptions 5. Comprehensive-cover complete domain

Deliberate falsification (unwillingness to respond accurately

1. Invasion of privacy 2. Time pressure and fatigue 3. Physical or social environment 4. Questionnaire-specific issues 5. Mischievous respondents 6. Social desirability bias

What are the relative strengths and weaknesses of kiosk interviews? Relative Strengths of Interactive Kiosk Surv

1. Novel to most potential respondents a. The current novelty of this data-collection approach may pique the interest of respondents who other wise would ignore the opportunity to participate in your study. 2. Provide nonaggressive solicitation Other forms of data collection require active solicitation of respondents through telephone, conventional mail, and email requests. In contrast. Participation in a kiosk based study is entirely respondent motivated. 1. Allow on-the-fly data analysis a. Although you can conduct a pilot study, most self-administered questionaries' are created, distributed, and eventually returned for analysis. As as result, you can't fix faulty or omitted questions spotted from early returns. In contrast, kiosks linked to the internet can transmit responses immediately for analysis. If those responses suggest questioner modifications, you can make the necessary changes and upload a new improved question. Relative Weaknesses of Interactive Kiosk Surveys 1. Cumbersome to set up and break down a. Just as banquet rooms don't construct themselves the kiosk materials used for data collection will need to be delivered set up and hauled off which can prove physical taxing. Also, venue kiosk- typically are reluctant to offer space floor and disturb traffic patterns. 1. Impersonal and unappealing a. Hard to attract people if unappealing

Constructing liker scales 3 efficient matrix organizations

1. Numeric matrix organization a. Scale numbers under each of the scale descriptions-1-5 2. Nonnumeric matrix organization a. Stongly agree, agree 3. Reversed numeric matrix organization a. Strongly agree 1, strongly agree 5

Two types of mail surveys

1. One shot or Cross-sectional survey a. Respondents receive a questionnaire with a cover letter and are asked to respond to that single questionnaire 2. Mail panel Respondent panel members receive mail surveys on a routine basis

Guidelines for writing good quesions

1. Only write questions that address your research problem 2. Write clear and precise questions 3. Include only mutually exclusive & exhaustive responses 4. Use natural and familiar language 5. Avoid leading questions 6. Ask one question at a time 7. soften the impact of potentially objectionable questions

Rules of Thumb For Writing Good Questions?

1. Only write questions that address your research problem 2. Write clear and precise questions a. How many standard-sized candy bars0 the roughly 2 oz. size you find in vending machines did you eat last week? None, 1-2,3-4 b. How often did you watch at least half of a major league baseball game in the last month? Never , once or twice, once a week. Natural language is intently imprecise. 3. Include only mutually exclusive & exhaustive responses a. Exhaustive means all possible responses are represented, and mutually exclusive means that each answer precludes all other answers. b. Exhaustive doesn't mean you should list every possible answer. Should be in most lkely choices. Capture rare respondents with other, please. 4. Use natural and familiar language a. Will collect more reliable data if you use regionally accepted terms in questions. Chicago not yall. 5. Avoid leading question a. What did you dislike about the product you just tried? b. It will make think about disliking it even if they haven't 6. ask one question at a time a. double barreled version- should the city use tax revue to build a new community swimming pool that includes lanes for swimming laps ut isn't enclosed for winter use. 7. soften the impact of potentially objectionable questions a. innocuous question- is the last digit of your social security number odd? b. Sensitive question- would you support a bill legalizing marijuana.

How Questions Should be Structured

1. Open ended. 2. Closed ended 1. Open ended- responded state answer in their own words 2. Close ended- respondents chose answer from predetermined options

reviewing the contemporary methods for conducting phone interviews How many types of phone interviews are there? 3 types of Face to Face surveys Phone interview methods

1. Paper-and-pencil administration a. Interviewer reads through the interview and records the respondents answers on paper. b. Eventually, those answers are entered into a computer file for data analysis. The additional data entry step- transferring written responses to a computer file- can introduce transcription errors, a type of administration error. 2. Computer-assisted administration a. Interviewer uses a computer for survey administration and data recording b. Is a more accurate and somewhat more sophisticated approach. It differs from paper-and-pencil administration in that interviewers use a computer for survey administration and data recording. Always get answers because of selection avoid impossible answers. 3. Computerized and Voice-activated interview a. Respondents speak to a machine b. NOT recommended c. Another way to conduct a telephone survey which we all dread. In essence, respondents speak to a machine, which is unlikely to garner thoughtful responses due to the interviews impersonal nature.

Strengths of mail interviews Relative weaknesses of mail interviews

1. Questionnaire can be lengthy if you use proper procedure and incentives a. Because respondents can complete a mail questionnaire at their convenience and over multiple sessions, it can be longer than other types of questionnaires. 2. Interviewer-induced response bias isn't a concern because interviewers are not present. a. Because their replies can be anonymous- and they may doubt a face to face or telephone interviewers promise of confidentiality- respondents are more likely to answer personal and controversial questions extensively and truthfully. 3. It's a low cost method, especially if your conducting a national survey. a. First-class postage is the same regardless of distance. Hence a national mail survey can be fielded at the same cost as a local mail survey. Relative weaknesses of mail interviews 1. They don't permit quick data collection a. You can indicate clear deadlines for questionnaire return, but the multi-mailing procedure characteristic of mail-based surveys means that you'll measure data-collection time in months rather than days or weeks. 2. Lack of human oversight can compromise data quality a. Poorly designed questionnaires will have low response rates, and people who choose to participate in your survey may answer questions selectively. Also, the potential for respondent errors is meaningful, especially in following your instructions for skipping questions that don't pertain to the respondent. 3. They preclude questioning versatility a. The highly standardized format can lead respondents to answer quickly or become bored with the process, in resulting in poor-quality data, especially if you include many scaled questions.

Reasons People Become Nonrespondents

1. Self-selection bias a. People may choose, because of some personal or study-related charactertic, not to participate in your survey. 2. Fear and anxiety 3. Invasion of privacy 4. Hostility towards a. Study sponsor b. Study topic c. interviewer

Breakdown of Basic Hierarchy-of-Effects Model

1. Step 1: People become aware that a product exists. a. Awareness can be measured through aided or unaided recall. TV commercial? 2. Step 2: People discover the attributes of that product. a. These attribute discoveries represent factual knowledge. 3. Step 3: People begin to form emotions toward it. a. People begin to like or dislike that product, and develop the affective component of attitude 4. Step 4: People may intend to buy it on the next purchase occasion. a. This step relates to the behavioral component of attitude. 5. Step 5: People act on their intentions. a. This is the culmination of the hierarchy-of-effects model, where predispositions lead to action

Breakdown of Basic Hierarchy-of-Effects Model

1. Step 1: People become aware that a product exists. a. Awareness can be measured through aided or unaided recall. TV commercial? 2. Step 2: People discover the attributes of that product. a. These attribute discoveries represent factual knowledge. 3. Step 3: People begin to form emotions toward it. a. People begin to like or dislike that product, and develop the affective component of attitude 4. Step 4: People may intend to buy it on the next purchase occasion. a. This step relates to the behavioral component of attitude. 5. Step 5: People act on their intentions. a. This is the culmination of the hierarchy-of-effects model, where predispositions lead to action empathizes that all three components of attitudes must be measured.

What are the relative strengths and weaknesses of face to face interviews?

1. They encourage respondent cooperation and discourage question nonresponse a. When respondents agree to participate in a face-to-face interview, they committed their time to assist you. This time commitment tends to ensure genuine and thoughtful responses to you questions. The same commitment also reduces the likelihood that respondents will refuse to answer reasonable questions that are consistent with your stated research goals. 2. They allow versatile and extensive questioning a. With these interviews you spend more time with each respondent, so you can show visual materials- such as pictures, drawings and videos- or demonstrate physical items like products. As a result, interviewers can explore consumer attitudes and preferences for alternative product configurations, packaging, advertising, and the like. 3. They produce high-quality data a. Miscomprehension is never impossible- as anyone who's attempted a serious conversation with a significant other can attest- but you'll often be able to spot when respondents fail to grasp a question. In reaction to superficial answers, you can ask probing questions that reveal deeply held attitudes and preferences along with seldom discussed behavior. Relative weaknesses of face to face interviews 1. They don't permit quick data collection a. Unless you have access to well-trained staff of interviewers, you cant schedule and conduct lengthy face to face interviews overnight, especially if respondents are geographically dispersed. Given the need to drive between locations 2. Their need for skilled interviewers drives up costs a. Good face to face interviewers must be intelligent. Articulate, and well trained. Such people must be payed adequately for their time. MOST expensive approach for collecting survey data. 3. They can be subject to response bias a. Despite confidentiality promises, respondents and their answers are known to interviewers. As a result, social desirability and other response biases can degrade response quality. Notwithstanding extensive training, interviewers subconscious body language and even the way they dress may sway responses into more conventional directions.

Advantages of Open-Ended Questions

1. Wide Range of Answers 2. Unexpected Responses Prompt Good Follow Up Questions

How many types of face to face interviews are there?

3 types of Face to face surveys 1. Home 2. Office 3. (mall) Intercept Face-to face interviews typically are scheduled in advance, which establishes an interview completion date for a given study. Because the time needed to analyze interviewees responses is reasonably predictable, you can anticipate accurately when you'll have study results

Q7. Tom is getting ready to analyze the data from a survey but is becoming increasingly concerned that the changes in responses over respondents to the brand preference measures may not be due to actual changes in the brand preferences of the respondents but due to the way the questions on brand preference were asked. Tom is concerned with A) reliability. B) validity. C) sensitivity. D) generalizability. E) all of the above.

A

Likert Scales

A Likert scale is a series of statements plus a scale that people respond to by expressing their level of positive or negative evaluation of each statement. The example shows one item of a Likert scale. Q10. The question "I love watching the ads during the Super Bowl: 1) Strongly disagree, 2) Disagree, 3) Neither agree nor disagree, 4) Agree, 5) Strongly agree" is an example of a <-blank-> question. Which response below best fills in the blank.

Q5. Compared to mail surveys, online surveys benefit market researchers through all of the following EXCEPT ________. A) relatively high response rates. B) faster data collection C) lower cost D) more visually appealing E) All the other responses are benefits of online surveys.

A)—Although the answers B through D are benefits of online or browser surveys, A is the exception. Not only are response rates not a benefit of online surveys, it is near impossible to compute a response rate for most internet surveys. In fact, a drawback of internet surveys is that the samples are not representative, which implies that there is no potential sample for most internet surveys, Note that lower cost for mail surveys is relative to face to face, not browser, surveys, and is relevant especially with national surveys.

Avoid leading questions

Avoid Double-Barreled Questions A) Are you satisfied with the cost and convenience of using your smart phone? (B1) Are you satisfied with the cost of using your smart phone? (B2) Are you satisfied with the convenience of using your smart phone

Writing Good Questions C R A U

Characteristics of good questions 1. Concise 2. Relevant to the research problem 3. Appealing 4. Unambiguous

What are interactive kiosk interviews?

Comparable to but far more technologically sophisticated than an electronic voting booth, an interactive kiosk provides a unique arena for capturing data. It can be set up near a facility's main entrance or in a central location. Data are captured via computer; this technological aspect of the process may prove appealing to potential respondents. 1. More sophisticated than an electronic voting booth 2. Provides a unique arena for capturing data 3. Can be set up near a central location 4. May prove technologically appealing to respondents

Self-Administered, Electronic Surveys how many types of self administered electronic interviews are there?

How many types of self-administered electronic interviews are there? 3 types of self administered electronic surveys 1. Internet browsers 2. E-Mail 3. Interactive Kiosks

hierarchy0 of0effects model

L F D awareness 1.aided 2.unaided knowledge (belief) 1. attribute assessment 2. perceptions Affect 1`.like./disklike 2. first choice 3. consideration set 4. intentions 5. behavior

How to boost response rates?

Minimizing error by boosting your response rates When response rates decline, the representativeness of survey results decline as well. Declining response rates are a major problem because they reduce research quality and increase data collection costs.

Face to Face

Most versatile most peronsal

Remember a questionnaire is only as good as its questions

Reject biased questions BIASED questions defeat the purpose of MR Good questions are unbiased. They should measure respondents' attitudes, preferences, and behaviors accurately

variation in responses

Researchers hope that variation in responses (measures) are only due to variation in the characteristic being measured, and not the reasons on the right 1. Stable characteristics of individual respondents 2. Short-term personal factors a. Once tired 3. Situational factors a. Peoples responses to questions differ by their immediate situation. 4. Variation in administering the questionnaire 5. Specific items included in the questionnaire 6. Lack of clarity in the measurement instrument a. What's your sex? Today what is your gender? 7. Mechanical or instrument factors

structing Likert scales

Reversed Coded Items 1. So the total sum of the scale is meaningful 2. From the highest code plus 1 subtract each code 3. For a 5-point scale: 5 1, 4 2, 3 remains unchanged, 2 4, and 1 5 1. Providing verbal category descriptions a. Clear and precise on meaning 2. Determining the number of response categories a. Up to 9? 3. Choosing a balanced or an unbalanced scale 4. Selecting odd or even number of categories a. Even get rid of fence sitters.

SD Scale Underlying Dimensions

SD Scale Underlying Dimensions 1. Evaluation 2. Power 3. Activity 1. Regardless of culture or language, uses to evaluate things in their social environment.

Telephone surveys

Telephone interviews can be used to contact respondents at work or at home; and with cellphones, respondents can be reached while traveling. ephone surveys 1. Call center with interviews manning each telephone and supervisor randomly monitoring interviews for quality control 2. Effective if questionnaire length reasonable and respondent are caught at convenient time 3. Prime method for collecting attitudinal, behavioral, and descriptive data.

Face- to - face interviews

To boost response rate for face to face interviews you can offer meaningful incentives, dress professionally, and solict potential respondents at the most convenient time of day,

What are the relative strengths and weaknesses of browser interviews?

What are the relative strengths and weaknesses of browser interviews? Relative strengths of Browser Surveys 1. Speedy data collection a. After you click the send button on your email invitation, potential respondents should receive that invitation within minutes. You'll likely receive completed questionnaires within hours. 2. Visually appealing and interactive 3. Powerful software infrastructure a. Because HTML and XML provide flexible software for creating questionnaires, you can present questions in a broad range of formats. 4. Cheaper distribution and processing costs Relative weakness of Browser surveys 1. Possible misunderstanding and/ or misinterpretation of questions are from zero. a. Despite your best efforts to design a clear questionnaire, respondents may misinterpret some of your instructions and questions. Without an interviewer present to sense such confusion, its likely you'll mistakenly include these meaningless responses in your data files and subsequent analyses. 2. Samples lack representativeness a. More than any other type of survey, browser-based surveys are subject to self-selection bias. In other words people who opt in typically are far more enthusiastic about the topic in question, far more willing to provide their opinions, and far more likely to complete. Lack internet. 3. Cooperation depends on how solicitation email is perceived. Spam 4. Respondent recontact and /or follow up difficult a. If the don't submit their email address how can you re contact them. Browser-based surveys require proactive respondents who will accept an invitation to participate and subsequently log on to a web site. In contras, email surveys merely require reactive participants who will answer the questions posed in your email and return them to you

email interviews str and weak

What are the relative strengths and weaknesses of e-mail interviews? One advantage of Email surveys 1. Rapid distribution & Collection 2. can be sent and returned quickly. All that's needed are a potential respondents email address, a brief introduction to your survey, a questionnaire, and a processes within a few days because fewer steps are required. Relative weaknesses of Email interviews 1. Limits question types and layouts a. Extensive differences in the capabilities of respondent's pcs, internet service providers, and email software limit the types of questions and questionnaire layouts to be used. 1. E-mails aren't secure, hacking can occur a. Unlike browser-based interactions, which can be safeguarded through well-established encryption procedures, most e-mail correspondence is unencrypted. As a result, identity thief's. 2. People reluctant to open e-mail attachments from unknown sources a. Hackers can embed spyware, viruses, Trojan horses, and other malware into email attachments. All that's necessary to install and activate such maleficent software is to pen the attachment.

Factors to Consider When Choosing a Data-Collection Method

What factors must be considered when choosing a type of survey? Considerations when choosing a data collection method 1. Available budget 2. How quickly data needed 3. Complexity/Structure/length of questionnaire 4. Need for stimuli exposure 5. Importance of a representative sample 6. Data quality 7. Respondent interest 8. Ease of obtaining respondent cooperation 9. Incidence rate

what is random sampling error and how can it be minimized.

What is (random) sampling error and how can it be minimized? Random Sampling Error-Statistical fluctuation that occurs because of chance variation in the elements selected for the sample. random sampling error definition. beyond your control

Distinguish between Symmetric and Non-Symmetric Scales and Balanced and Unbalanced Ones?

When the response scale contains an equal number of positive and negative responses, the scale is said to be SYMMETRIC or BALANCED When the response scale contains an unequal number of positive and negative responses, or only the positive or negative responses, the scale is said to be NON- SYMMETRIC or UNBALANCED

nonresponse bias

a difference between those who do and do to respond

inability to respond accurately

a. To an extent, articulateness is associated with the ability to respond accurately, and some people are more articulate than others. 2. Ignorance a. Just because you ask them a question doesn't mean they will know the answer. 1. They can't answer knowledgably about other people 2. they cant predict their own behavior well

internet samples

although online surveys tend to rely on nonscientific opt in samples, Internet samples may still be representative of a target population and allow you to access hard- to- reach respondents.

reliabliilty

degree to which measures are free from random error and, therefore, provide consistent data. The extent to which the survey responses are internally consisten

validity

degree to which what the researcher was trying to measure was actually measured

Identifying Nonresponse Error

less than 100 percent response rate; nonresponse error

can something be valid without being reliable?

no

Q9. <-blank-> questions allow respondents to answer in an uncountable number of ways and often reveal more about how people think.? A) Dichotomous. B) Likert scale. C) Open-ended. D) Multiple-choice. E) All of the other answers are correct.

open ended

Q9. <-blank-> questions allow respondents to answer in an uncountable number of ways and often reveal more about how people think.? 1. A) Dichotomous. 2. B) Likert scale. 3. C) Open-ended. 4. D) Multiple-choice. 5. E) All of the other answers are correct.

open ended

Self-Administered, Paper-and-Pencil Surveys

paper and pencil (more thought) and electronic (quicker)

response bias

people answer a question incorrectly

least personal

self-administers questionnaire

Q8. A scale that connects words that are extremities, if not complete opposites, along a scale is called a <-blank-> question. Which <-blank-> best fills in the answer? A) dichotomous. B) Likert scale. C) open-ended. D) multiple-choice. E) semantic differential.

semantic differntial

systematic error sources 1

systematic error sources 1. Sampling frame 2. Respondent error 3. Administrative

test reliabliry

test and retest, stability, equivalent form, internal consistency, split half technique

Nonresponse Error

• Bias that results when the people who choose not to respond differ systematically from those who choose to respond

what are the relative strengths and weaknesses of intercept interviews?

• Called mall intercept because they tend to take place in malls, but could take place anywhere that people gather • Usually conducted by professionals, who have permission from lieu (mall) owners. • Could take place in a dedicated office with audiovisual equipment. Incentives often given to respondents Relative strengths of intercept interviews 1. They make relatively speedy data collection possible a. Assuming you have a popular location, multiple interviewers, and a reasonable-length questionnaire, you can complete hundreds of interviews in a few days. 2. They make versatile and extensive questioning possible a. As with face-to-face interviews in general, intercept interviews allow you to show visual materials or demonstrate physical items. In particular, they are especially well suited to testing foods and evaluating ads. The interactivity of such interviews can boost response rate rates Relative weaknesses of intercept interviews 1. Geographically inflexible a. Due to urban and local biases, your findings may not pertain to rural settings or distant regions. Relying on shoppers at a single local mall to help you test a regional or national marketing campaign is problematic. 1. Respondent cooperation is moderate at best a. Artifacts of this moderate cooperation level include limited questionnaire length and a higher number of incomplete interviews. Because most consumers visit malls to shop, they'll resist completing long and time-consuming questionnaires, which limits the number and types of questions you can ask each respondent. For example, respondents may ignore lengthy scenarios 2. Callbacks or follow-ups are difficult a. Respondents may be reluctant to provide contact information (for fear of future solicitation), which limits your ability to verify answers later. Also, willingness to participate in an intercept interview is a momentary decision incompatible with the commitment required for participation in longitudinal studies.

What is systematic survey error and what are its sources? (also called nonsampiling errors)

• Caused by some imperfect aspect of the research design or a mistake in research execution Are unrepresentative sampling results due to study design or execution flaws


Set pelajaran terkait

Criminal Justice: Chapter 15 Test

View Set

NU322 - Ticket to Test Acid Base

View Set

H.A in Nursing Ch19 Assessing Thorax and Lungs

View Set

Терміни з історії. ЗНО. Тема 5

View Set

Chapter 10 Operations Management

View Set

Primary and Secondary Immunodeficiency

View Set