Chapter 10 & 13,14,15 DC's

What is the purpose of questionnaires and observation forms?

-2 basic methods for obtaining quant primary data in descriptive research -procedure for standardizing the data-collection process so that the data obtained are internally consistent and can be analyzed in a uniform and coherent manner -this will ensure compatibility of the data, increase speed and accuracy of recording, and facilitate data processing

What is the proper order for questions intended to obtain basic, classification, and identification information?

-Basic info relates directly to research problem-taken first bc greatest importance -Classification info: socioeconomic and demographic characteristics used to classify respondents and understand results -Identification info: name, postal address, email, phone number -difficult/sensitive questions should be placed late in the sequence such as income at end of classification and phone number at end of identification -funnel approach-general to specific

What are the advantages and disadvantages of unstructured questions?

-adv: good first questions on a topic, enable respondents to express general attitudes and opinions that can help researcher interpret responses to structured questions,less biasing influence on response, provide rich insights, useful for explanatory research -dis: potential for interviewer bias is high (data depends on interviewer's skill), coding is costly and time consuming, give weight to respondents that are more articulate, not suitable for self or computer assisted questionnaires

What guidelines are available for deciding on the form and layout of a questionnaire?

-divide questionnaire into several parts -questions in each part should be numbered -questionnaire should be pre-coded -questionnaire should be numbered serially

What can a researcher do to make the request for information seem legitimate?

-explain why the data is needed -what is the purpose of the data

What are some of limitations of cross tabulations?

-more than 3 variables can be cross tabbed but interpretation is complex -should be at least five expected observations in each cell for the statistics computed to be reliable -inefficient way of examining relationships when serval variables -cross tabs examines associations between variables NOT causation

Explain how the mode of administration affects questionnaire design.

-personal interview: see questionnaire and interact face-to-face with interviewer so can use lengthy, complex and varied questions-conversational style -telephone: interact with interviewer but cannot see questionnaire so use simple, short questions-conversational style -mail: self-administered so simple questions with detailed instructions -computer assisted interviewing (CATI/CAPI): complex skip patterns and randomization of questions can eliminate order bias -Internet and Email: simpler than mail questionnaires

Describe the data-preparation process.

-preparing preliminary plan of data analysis: formulated in research design phase (how process is guided) -questionnaire checking -editing -coding -transcribing -data cleaning (treatment for missing responses) -statistically adjusting the data (make representative of population of interest) -selecting a data analysis strategy (differs from preliminary phase due to introduction of new insights and info) *Data prep should begin as soon as the 1st batch of questionnaires are taken from field and while the fieldwork is going on to potentially modify and incorporate corrective action

What aspects are involved in the supervision of fieldworkers?

-quality control and editing -sampling control -control of cheating -central office control

Explain the use of randomized techniques in obtaining sensitive information.

-respondents are presented with two questions (one sensitive, one neutral)with a known probability of a "yes" response -asked to select one question randomly (ex. flip a coin) - answers the question "yes" or "no" without telling researcher which one answering -researcher can look at the probabilities of a "yes" to overall probability, probability of choosing the sensitive question, and probability of a "Yes" response to the neutral question -can determine P(yes for sensitive) using law of probability but cannot determine which respondents answered

How are unsatisfactory responses discovered in editing treated?

-returning to the field (recontact with the respondents, good for when small identifiable sample) -assign missing values (desirable if the number of respondents with unsatisfactory responses is small, the proportion of unsat. responses is small, or the variables with the unsat. responses are not the key variables) -discarding unsatisfactory respondents (may bias and should be reported- when desirable pg 422)

Discuss the reasons for the frequent use of cross-tabulations. What are some of its limitations?

Widely used in commercial marketing research because: 1. cross tab analysis and results can be easily interpreted and understood by managers who aren't statistically oriented 2. the clarity of interpretation provides a stronger link between the research results and managerial action 3. a series of cross tabs may provide greater insights into a complex phenomenon than a single multivariate analysis 4. cross tab may alleviate the problem of sparse cells, which could be serious in discrete multivariate analysis 5. cross tab analysis is simple to conduct and appealing to less sophisticated researchers

What is a double-barreled question?

a single question that attempts to cover two issues. Such questions can be confusing to respondents and result in ambiguous responses

CH10 Problem 7-Critique questionnairre

incorrectly obtains identification (name) and some classification (demographic) info at the beginning

Describe the issues involved in pretesting a questionnaire.

- the respondents should be similar to those included in actual survey-- drawn from same population -best done by personal interviews (even if using other method) to gauge reaction and attitudes then use methods of actual survey -administer questionnaire in an environment and context similar to actual survey -sample size of respondents increase substantially if the pretesting involves several stages or waves -after each significant revision of questionnaire, another pretest should be conducted using different sample - sound pretesting involves several stages -pretesting continues until no further stages needed

What are the issues involved in designing multiple choice questions?

-the number of alternatives(all possible choices and mutually exclusive) -it is difficult to obtain info on alternatives not listed -even with "Other" category, respondents tend to choose among listed alternatives - biased responses from list of alternatives -potential for order bias

Explain the concepts of aided and unaided recall.

-unaided: questions that do not provide the respondent with cues to the event-- can underestimate the actual; occurrence of an event -aided: attempts to stimulate the respondent's memory by providing cues related to the event of interest-- guard against biasing the responses by too much stimulation

What is the validation of fieldwork?How is this done?

-verifying that fieldworkers are submitting authentic interviews -call 10-25% of respondents to make sure interviews were conducted -ask about length and quality of interview, reaction to interviewer, and basic demographic data (cross checked by reported info)

What are the guidelines available on deciding on question wording?

1. define the issue 2. use ordinary words 3. use unambiguous words 4. avoid leading questions 5. avoid implicit alternatives 6. avoid implicit assumptions 7. avoid generalizations and estimates 8. use positive and negative statements

Describe the fieldwork/data-collection process.

1. selection of fieldworkers -select based off specifications/characteristics, mode of collection, and type of respondent 2. training -ensure uniformity in administration -initial contact, probe, questions, recording answers, and terminating interview 3. supervision -obtain periodic reports 4. validation/verification of fieldwork 5. evaluation -provide performance feedback

What is the difference between pre-coding and post-coding?

coding def: the assignment of a code to represent a specific response to a specific question along with the data record and column position that code will occupy -pre-coded: only structured questions or very few unstructured questions- the codes are assigned before fieldwork is assigned -post-coded: unstructured questions-codes are assigned after the questionnaires have been returned from the field

What kind of consistency checks are made in cleaning then data?

data cleaning def: thorough and extensive checks for consistency and treatment of missing responses -consistency checks: identify data that are out of range, logically inconsistent, or have extreme values. Data with values not defined by the coding scheme are inadmissible

What is a leading question? Give an example.

def: a question that gives the respondent a clue as to what answer is desired or leads the respondent to answer in a certain way ex. Do you believe TSwift was right to get mad at Kanye West for the song Famous?

What is meant by editing a questionnaire?

def: a review of the questionnaires with the objective of increasing accuracy and precision of the collected data -screening questions to identify illegible (poorly recorded responses from unstructured questions or if they aren't properly coded), incomplete, inconsistent, or ambiguous responses

What options are available for the treatment of missing data?

def: values of a variable that are unknown, because the respondents did not provide unambiguous answers to the question (ambiguous responses or poorly coded) 1. Substitute a neutral value (mean response) 2. Substitute an imputed response (infer from available data what it would be) 3. Casewise deletion (cases/respondents are discarded from the analysis) 4. Pairwise deletion (use only the cases/respondents with complete responses for the variable involved in each calculation rather than throw it out completely)

What is the major difference between cross-tabulation and frequency distribution?

major difference: a frequency distribution describes one variable at a time and a cross tabulation describes two or more variables simultaneously (the merging of a freq dist of two or more variables on a single table to check association or relationships) -Cross tabs (contingency table) results in tables that reflect the joint distribution of 2+ variables that have a limited number of categories or distinct values (nominal scale)

What are the reasons that respondents are unable to answer the question asked?

respondent may not be informed, may not remember, or may be unable to articulate

What is a skewed distribution? What does it mean?

skewness def: a characteristic of a distribution that assesses its symmetry from the mean--tendency of the deviations from the mean to be larger in one direction than in the other (one tail heavier). -the positive (right) and negative (left) directions are unequal (not centered around the mean)

What are the reasons that respondents are unwilling to answer specific questions?

too much effort is required, the situation or context may not seem appropriate for disclosure, no legitimate purpose or need for the information requested is apparent, or the information requested is sensitive