• No results found

Methods: Data Collection & Survey Design

N/A
N/A
Protected

Academic year: 2021

Share "Methods: Data Collection & Survey Design"

Copied!
203
0
0

Loading.... (view fulltext now)

Full text

(1)Methods: Data Collection & Survey Design Dr. Patrick Toche. References :. † William G. Zikmund, Barry J. Babin, Jon C. Carr, Mitch Griffin, Business Research Methods, South-Western College Pub, 9th edition (2012), 9781111826925. Other references may be given from time to time.. 1 / 84.

(2) Learning Objectives. 2 / 84.

(3) Learning Objectives 1. Outline the stages of the research process. 2. Understand the ethical issues raised by research. 3. Distinguish the various categories of surveys. Describe the category of questions used in a questionnaire. 4. Understand the principles of experimental design. Identify sampling errors and systemic biases. 5. Understand the purpose of case studies, experience surveys, and focus group interviews. Distinguish primary and secondary data.. 3 / 84.

(4) Learning Objectives 1. Outline the stages of the research process. 2. Understand the ethical issues raised by research. 3. Distinguish the various categories of surveys. Describe the category of questions used in a questionnaire. 4. Understand the principles of experimental design. Identify sampling errors and systemic biases. 5. Understand the purpose of case studies, experience surveys, and focus group interviews. Distinguish primary and secondary data.. 3 / 84.

(5) Learning Objectives 1. Outline the stages of the research process. 2. Understand the ethical issues raised by research. 3. Distinguish the various categories of surveys. Describe the category of questions used in a questionnaire. 4. Understand the principles of experimental design. Identify sampling errors and systemic biases. 5. Understand the purpose of case studies, experience surveys, and focus group interviews. Distinguish primary and secondary data.. 3 / 84.

(6) Learning Objectives 1. Outline the stages of the research process. 2. Understand the ethical issues raised by research. 3. Distinguish the various categories of surveys. Describe the category of questions used in a questionnaire. 4. Understand the principles of experimental design. Identify sampling errors and systemic biases. 5. Understand the purpose of case studies, experience surveys, and focus group interviews. Distinguish primary and secondary data.. 3 / 84.

(7) Learning Objectives 1. Outline the stages of the research process. 2. Understand the ethical issues raised by research. 3. Distinguish the various categories of surveys. Describe the category of questions used in a questionnaire. 4. Understand the principles of experimental design. Identify sampling errors and systemic biases. 5. Understand the purpose of case studies, experience surveys, and focus group interviews. Distinguish primary and secondary data.. 3 / 84.

(8) By a small sample, we may judge of the whole piece. Miguel de Cervantes, The History of the Renowned Don Quixote de la Mancha, Part I, Book I, chapter 4, Robert & Andrew Foulis, 1771, “translated by several hands”. In Spanish (1605): “por el hilo se sacará el ovillo” — pull on the thread to unspool the coil.. It is a capital mistake to theorize before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. Arthur Conan Doyle, The Adventures of Sherlock Holmes (1892).. If your experiment needs a statistician, you need a better experiment. attributed to Ernest Rutherford (1871-1937), “founder” of nuclear physics.. 4 / 84.

(9) Research Process. 5 / 84.

(10) The Research Process I Interpret empirical evidence to confirm or disprove a hypothesis. I Stages in the research process: 1. Research objectives. 2. Research design. 3. Sample design. 4. Data collection. 5. Data analysis. 6. Research report. 1. Exploratory Research: Define the nature of a problem, clarify ambiguous situations, discover potential opportunities, shapes hypotheses, outline research strategy. 2. Descriptive Research: Describes the characteristics of objects, people, groups, organizations, or environments. Who? What? When? Where? Why? How? 3. Causal Research: Identify cause and effect. Rule out spurious correlations.. 6 / 84.

(11) The Research Process I Interpret empirical evidence to confirm or disprove a hypothesis. I Stages in the research process: 1. Research objectives. 2. Research design. 3. Sample design. 4. Data collection. 5. Data analysis. 6. Research report. 1. Exploratory Research: Define the nature of a problem, clarify ambiguous situations, discover potential opportunities, shapes hypotheses, outline research strategy. 2. Descriptive Research: Describes the characteristics of objects, people, groups, organizations, or environments. Who? What? When? Where? Why? How? 3. Causal Research: Identify cause and effect. Rule out spurious correlations.. 6 / 84.

(12) The Research Process I Interpret empirical evidence to confirm or disprove a hypothesis. I Stages in the research process: 1. Research objectives. 2. Research design. 3. Sample design. 4. Data collection. 5. Data analysis. 6. Research report. 1. Exploratory Research: Define the nature of a problem, clarify ambiguous situations, discover potential opportunities, shapes hypotheses, outline research strategy. 2. Descriptive Research: Describes the characteristics of objects, people, groups, organizations, or environments. Who? What? When? Where? Why? How? 3. Causal Research: Identify cause and effect. Rule out spurious correlations.. 6 / 84.

(13) The Research Process I Interpret empirical evidence to confirm or disprove a hypothesis. I Stages in the research process: 1. Research objectives. 2. Research design. 3. Sample design. 4. Data collection. 5. Data analysis. 6. Research report. 1. Exploratory Research: Define the nature of a problem, clarify ambiguous situations, discover potential opportunities, shapes hypotheses, outline research strategy. 2. Descriptive Research: Describes the characteristics of objects, people, groups, organizations, or environments. Who? What? When? Where? Why? How? 3. Causal Research: Identify cause and effect. Rule out spurious correlations.. 6 / 84.

(14) The Research Process I Interpret empirical evidence to confirm or disprove a hypothesis. I Stages in the research process: 1. Research objectives. 2. Research design. 3. Sample design. 4. Data collection. 5. Data analysis. 6. Research report. 1. Exploratory Research: Define the nature of a problem, clarify ambiguous situations, discover potential opportunities, shapes hypotheses, outline research strategy. 2. Descriptive Research: Describes the characteristics of objects, people, groups, organizations, or environments. Who? What? When? Where? Why? How? 3. Causal Research: Identify cause and effect. Rule out spurious correlations.. 6 / 84.

(15) Experimental Design. 16 / 84.

(16) Experimental Design Controlled experiment The experimenter manipulates one or more independent variables and holds constant all other independent variables while observing the effect on the dependent variable. In quasiexperiments, natural conditions that influence the variation are selected for observation.. I Experimental design:. – – – – I Experimental checks: – – –. Manipulation of the independent variables. Selection and measurement of the dependent variables. Selection and assignment of experimental subjects. Control over extraneous variables. Validity Reliability Replicability. 17 / 84.

(17) Experimental Design Controlled experiment The experimenter manipulates one or more independent variables and holds constant all other independent variables while observing the effect on the dependent variable. In quasiexperiments, natural conditions that influence the variation are selected for observation.. I Experimental design:. – – – – I Experimental checks: – – –. Manipulation of the independent variables. Selection and measurement of the dependent variables. Selection and assignment of experimental subjects. Control over extraneous variables.. Validity Reliability Replicability. 17 / 84.

(18) Experimental Design Controlled experiment The experimenter manipulates one or more independent variables and holds constant all other independent variables while observing the effect on the dependent variable. In quasiexperiments, natural conditions that influence the variation are selected for observation.. I Experimental design:. – – – – I Experimental checks: – – –. Manipulation of the independent variables. Selection and measurement of the dependent variables. Selection and assignment of experimental subjects. Control over extraneous variables.. Validity Reliability Replicability. 17 / 84.

(19) Experimental Design Controlled experiment The experimenter manipulates one or more independent variables and holds constant all other independent variables while observing the effect on the dependent variable. In quasiexperiments, natural conditions that influence the variation are selected for observation.. I Experimental design:. – – – – I Experimental checks: – – –. Manipulation of the independent variables. Selection and measurement of the dependent variables. Selection and assignment of experimental subjects. Control over extraneous variables.. Validity Reliability Replicability. 17 / 84.

(20) Experimental Design Controlled experiment The experimenter manipulates one or more independent variables and holds constant all other independent variables while observing the effect on the dependent variable. In quasiexperiments, natural conditions that influence the variation are selected for observation.. I Experimental design:. – – – – I Experimental checks: – – –. Manipulation of the independent variables. Selection and measurement of the dependent variables. Selection and assignment of experimental subjects. Control over extraneous variables.. Validity Reliability Replicability. 17 / 84.

(21) Experimental Design Controlled experiment The experimenter manipulates one or more independent variables and holds constant all other independent variables while observing the effect on the dependent variable. In quasiexperiments, natural conditions that influence the variation are selected for observation.. I Experimental design:. – – – – I Experimental checks: – – –. Manipulation of the independent variables. Selection and measurement of the dependent variables. Selection and assignment of experimental subjects. Control over extraneous variables.. Validity Reliability Replicability. 17 / 84.

(22) Experimental Design Controlled experiment The experimenter manipulates one or more independent variables and holds constant all other independent variables while observing the effect on the dependent variable. In quasiexperiments, natural conditions that influence the variation are selected for observation.. I Experimental design:. – – – – I Experimental checks: – – –. Manipulation of the independent variables. Selection and measurement of the dependent variables. Selection and assignment of experimental subjects. Control over extraneous variables.. Validity Reliability Replicability. 17 / 84.

(23) Experimental Design Controlled experiment The experimenter manipulates one or more independent variables and holds constant all other independent variables while observing the effect on the dependent variable. In quasiexperiments, natural conditions that influence the variation are selected for observation.. I Experimental design:. – – – – I Experimental checks: – – –. Manipulation of the independent variables. Selection and measurement of the dependent variables. Selection and assignment of experimental subjects. Control over extraneous variables.. Validity Reliability Replicability. 17 / 84.

(24) Experimental Design Controlled experiment The experimenter manipulates one or more independent variables and holds constant all other independent variables while observing the effect on the dependent variable. In quasiexperiments, natural conditions that influence the variation are selected for observation.. I Experimental design:. – – – – I Experimental checks: – – –. Manipulation of the independent variables. Selection and measurement of the dependent variables. Selection and assignment of experimental subjects. Control over extraneous variables.. Validity Reliability Replicability. 17 / 84.

(25) Experimental Design: The Control Group. http://dilbert.com/strip/2011-08-14/ Dilbert, August 14th, 2011.. 18 / 84.

(26) Experiments: Treatment & Control I Manipulation of the independent variable:. – – – – –. An experimental treatment refers to the way an experiment variable is manipulated. The treatment levels should be noticeably different and realistic. An experimental group refers to a group of subjects to whom an experimental treatment is administrated. A control group refers to a group of subjects to whom no experimental treatment is administrated. A placebo treatment is designed to appear like treatment but to have no real effect. The purpose of the placebo group is to account for the placebo effect, that is, effects from treatment that do not depend on the treatment itself.. I Selection and measurement of the dependent variable:. –. The dependent variables must represent an outcome of interest.. 19 / 84.

(27) Experiments: Treatment & Control I Manipulation of the independent variable:. – – – – –. An experimental treatment refers to the way an experiment variable is manipulated. The treatment levels should be noticeably different and realistic. An experimental group refers to a group of subjects to whom an experimental treatment is administrated. A control group refers to a group of subjects to whom no experimental treatment is administrated. A placebo treatment is designed to appear like treatment but to have no real effect. The purpose of the placebo group is to account for the placebo effect, that is, effects from treatment that do not depend on the treatment itself.. I Selection and measurement of the dependent variable:. –. The dependent variables must represent an outcome of interest.. 19 / 84.

(28) Experiments: Treatment & Control I Manipulation of the independent variable:. – – – – –. An experimental treatment refers to the way an experiment variable is manipulated. The treatment levels should be noticeably different and realistic. An experimental group refers to a group of subjects to whom an experimental treatment is administrated. A control group refers to a group of subjects to whom no experimental treatment is administrated. A placebo treatment is designed to appear like treatment but to have no real effect. The purpose of the placebo group is to account for the placebo effect, that is, effects from treatment that do not depend on the treatment itself.. I Selection and measurement of the dependent variable:. –. The dependent variables must represent an outcome of interest.. 19 / 84.

(29) Experiments: Treatment & Control I Manipulation of the independent variable:. – – – – –. An experimental treatment refers to the way an experiment variable is manipulated. The treatment levels should be noticeably different and realistic. An experimental group refers to a group of subjects to whom an experimental treatment is administrated. A control group refers to a group of subjects to whom no experimental treatment is administrated. A placebo treatment is designed to appear like treatment but to have no real effect. The purpose of the placebo group is to account for the placebo effect, that is, effects from treatment that do not depend on the treatment itself.. I Selection and measurement of the dependent variable:. –. The dependent variables must represent an outcome of interest.. 19 / 84.

(30) Experiments: Treatment & Control I Manipulation of the independent variable:. – – – – –. An experimental treatment refers to the way an experiment variable is manipulated. The treatment levels should be noticeably different and realistic. An experimental group refers to a group of subjects to whom an experimental treatment is administrated. A control group refers to a group of subjects to whom no experimental treatment is administrated. A placebo treatment is designed to appear like treatment but to have no real effect. The purpose of the placebo group is to account for the placebo effect, that is, effects from treatment that do not depend on the treatment itself.. I Selection and measurement of the dependent variable:. –. The dependent variables must represent an outcome of interest.. 19 / 84.

(31) Experiments: Treatment & Control I Manipulation of the independent variable:. – – – – –. An experimental treatment refers to the way an experiment variable is manipulated. The treatment levels should be noticeably different and realistic. An experimental group refers to a group of subjects to whom an experimental treatment is administrated. A control group refers to a group of subjects to whom no experimental treatment is administrated. A placebo treatment is designed to appear like treatment but to have no real effect. The purpose of the placebo group is to account for the placebo effect, that is, effects from treatment that do not depend on the treatment itself.. I Selection and measurement of the dependent variable:. –. The dependent variables must represent an outcome of interest.. 19 / 84.

(32) Experiments: Treatment & Control I Manipulation of the independent variable:. – – – – –. An experimental treatment refers to the way an experiment variable is manipulated. The treatment levels should be noticeably different and realistic. An experimental group refers to a group of subjects to whom an experimental treatment is administrated. A control group refers to a group of subjects to whom no experimental treatment is administrated. A placebo treatment is designed to appear like treatment but to have no real effect. The purpose of the placebo group is to account for the placebo effect, that is, effects from treatment that do not depend on the treatment itself.. I Selection and measurement of the dependent variable:. –. The dependent variables must represent an outcome of interest.. 19 / 84.

(33) Experiments: Treatment & Control I Manipulation of the independent variable:. – – – – –. An experimental treatment refers to the way an experiment variable is manipulated. The treatment levels should be noticeably different and realistic. An experimental group refers to a group of subjects to whom an experimental treatment is administrated. A control group refers to a group of subjects to whom no experimental treatment is administrated. A placebo treatment is designed to appear like treatment but to have no real effect. The purpose of the placebo group is to account for the placebo effect, that is, effects from treatment that do not depend on the treatment itself.. I Selection and measurement of the dependent variable:. –. The dependent variables must represent an outcome of interest.. 19 / 84.

(34) Experiments: Random Assignment I Selection and assignment: The participants selected for the experiment should be representative of the population of interest. Imagine you want to test a new drug and ask sick patients to volunteer. The problem is that the self-selection process may induce a bias, e.g. the patients who volunteer may typically be in poorer health than the average patient and therefore not ‘representative’.. I Systematic error: Occurs if the sampling units in an experimental cell are different from the units in another cell, and this difference affects the dependent variable. To control for systematic error, randomization and matching can be used.. I Randomization: Random assignment of treatment to groups. I Matching: Assigning participants in a way that their characteristics are the same each group.. 20 / 84.

(35) Experiments: Random Assignment I Selection and assignment: The participants selected for the experiment should be representative of the population of interest. Imagine you want to test a new drug and ask sick patients to volunteer. The problem is that the self-selection process may induce a bias, e.g. the patients who volunteer may typically be in poorer health than the average patient and therefore not ‘representative’.. I Systematic error: Occurs if the sampling units in an experimental cell are different from the units in another cell, and this difference affects the dependent variable. To control for systematic error, randomization and matching can be used.. I Randomization: Random assignment of treatment to groups. I Matching: Assigning participants in a way that their characteristics are the same each group.. 20 / 84.

(36) Experiments: Random Assignment I Selection and assignment: The participants selected for the experiment should be representative of the population of interest. Imagine you want to test a new drug and ask sick patients to volunteer. The problem is that the self-selection process may induce a bias, e.g. the patients who volunteer may typically be in poorer health than the average patient and therefore not ‘representative’.. I Systematic error: Occurs if the sampling units in an experimental cell are different from the units in another cell, and this difference affects the dependent variable. To control for systematic error, randomization and matching can be used.. I Randomization: Random assignment of treatment to groups. I Matching: Assigning participants in a way that their characteristics are the same each group.. 20 / 84.

(37) Experiments: Random Assignment I Selection and assignment: The participants selected for the experiment should be representative of the population of interest. Imagine you want to test a new drug and ask sick patients to volunteer. The problem is that the self-selection process may induce a bias, e.g. the patients who volunteer may typically be in poorer health than the average patient and therefore not ‘representative’.. I Systematic error: Occurs if the sampling units in an experimental cell are different from the units in another cell, and this difference affects the dependent variable. To control for systematic error, randomization and matching can be used.. I Randomization: Random assignment of treatment to groups. I Matching: Assigning participants in a way that their characteristics are the same each group.. 20 / 84.

(38) Experiments: Random Assignment I Selection and assignment: The participants selected for the experiment should be representative of the population of interest. Imagine you want to test a new drug and ask sick patients to volunteer. The problem is that the self-selection process may induce a bias, e.g. the patients who volunteer may typically be in poorer health than the average patient and therefore not ‘representative’.. I Systematic error: Occurs if the sampling units in an experimental cell are different from the units in another cell, and this difference affects the dependent variable. To control for systematic error, randomization and matching can be used.. I Randomization: Random assignment of treatment to groups. I Matching: Assigning participants in a way that their characteristics are the same each group.. 20 / 84.

(39) Experiments: Control I Control: Extraneous variables influence the relationship between the variables of interest in an experiment. These variables are undesirable because they add error to an experiment. Extraneous variables can produce confounded results.. I Confound: An alternative explanation, beyond the experimental variables, for observed differences in the dependent variable. A confound challenges the validity of the experiment.. I Internal validity: The extent that an experimental variable is truly responsible for any variance in the dependent variable.. I External validity: The accuracy with which experimental results can be generalized beyond the experimental subjects.. 21 / 84.

(40) Experiments: Control I Control: Extraneous variables influence the relationship between the variables of interest in an experiment. These variables are undesirable because they add error to an experiment. Extraneous variables can produce confounded results.. I Confound: An alternative explanation, beyond the experimental variables, for observed differences in the dependent variable. A confound challenges the validity of the experiment.. I Internal validity: The extent that an experimental variable is truly responsible for any variance in the dependent variable.. I External validity: The accuracy with which experimental results can be generalized beyond the experimental subjects.. 21 / 84.

(41) Experiments: Control I Control: Extraneous variables influence the relationship between the variables of interest in an experiment. These variables are undesirable because they add error to an experiment. Extraneous variables can produce confounded results.. I Confound: An alternative explanation, beyond the experimental variables, for observed differences in the dependent variable. A confound challenges the validity of the experiment.. I Internal validity: The extent that an experimental variable is truly responsible for any variance in the dependent variable.. I External validity: The accuracy with which experimental results can be generalized beyond the experimental subjects.. 21 / 84.

(42) Experiments: Control I Control: Extraneous variables influence the relationship between the variables of interest in an experiment. These variables are undesirable because they add error to an experiment. Extraneous variables can produce confounded results.. I Confound: An alternative explanation, beyond the experimental variables, for observed differences in the dependent variable. A confound challenges the validity of the experiment.. I Internal validity: The extent that an experimental variable is truly responsible for any variance in the dependent variable.. I External validity: The accuracy with which experimental results can be generalized beyond the experimental subjects.. 21 / 84.

(43) Experiments: Observer Bias I Observer bias: Errors associated with observations. Observers may accidentally introduce a bias by their presence alone — e.g. the Placebo effect, the Pygmalion effect, the Hawthorne effect.. I Placebo effect: Patients’ belief in the efficacy of a medical intervention can improve their health. A randomly assigned placebo group is given an inert drug (sugar in water) to control for the placebo effect.. I Pygmalion effect: Teachers’ expectations can strongly affect student performance — a type of self-fulfilling prophecy.. 22 / 84.

(44) Experiments: Observer Bias I Observer bias: Errors associated with observations. Observers may accidentally introduce a bias by their presence alone — e.g. the Placebo effect, the Pygmalion effect, the Hawthorne effect.. I Placebo effect: Patients’ belief in the efficacy of a medical intervention can improve their health. A randomly assigned placebo group is given an inert drug (sugar in water) to control for the placebo effect.. I Pygmalion effect: Teachers’ expectations can strongly affect student performance — a type of self-fulfilling prophecy.. 22 / 84.

(45) Experiments: Observer Bias I Observer bias: Errors associated with observations. Observers may accidentally introduce a bias by their presence alone — e.g. the Placebo effect, the Pygmalion effect, the Hawthorne effect.. I Placebo effect: Patients’ belief in the efficacy of a medical intervention can improve their health. A randomly assigned placebo group is given an inert drug (sugar in water) to control for the placebo effect.. I Pygmalion effect: Teachers’ expectations can strongly affect student performance — a type of self-fulfilling prophecy.. 22 / 84.

(46) Experiments: Hawthorne Effect I Hawthorne effect: Subjects perform differently when they know they are experimental subjects.. I In a study of the effects of various working conditions on productivity — such as hours of work, rest periods, lighting, and methods of pay — at the Western Electric Hawthorne plant in Cicero, Illinois, researchers found that workers’ productivity increased whether the work hours were lengthened or shortened, whether lighting was very bright or very dim, etc... I Productivity increased because workers reacted positively to being part of an experiment. This unintended effect is now known as the Hawthorne effect.. 23 / 84.

(47) Experiments: Hawthorne Effect I Hawthorne effect: Subjects perform differently when they know they are experimental subjects.. I In a study of the effects of various working conditions on productivity — such as hours of work, rest periods, lighting, and methods of pay — at the Western Electric Hawthorne plant in Cicero, Illinois, researchers found that workers’ productivity increased whether the work hours were lengthened or shortened, whether lighting was very bright or very dim, etc... I Productivity increased because workers reacted positively to being part of an experiment. This unintended effect is now known as the Hawthorne effect.. 23 / 84.

(48) Experiments: Hawthorne Effect I Hawthorne effect: Subjects perform differently when they know they are experimental subjects.. I In a study of the effects of various working conditions on productivity — such as hours of work, rest periods, lighting, and methods of pay — at the Western Electric Hawthorne plant in Cicero, Illinois, researchers found that workers’ productivity increased whether the work hours were lengthened or shortened, whether lighting was very bright or very dim, etc... I Productivity increased because workers reacted positively to being part of an experiment. This unintended effect is now known as the Hawthorne effect.. 23 / 84.

(49) Occam’s Razor Occam’s Razor Principle: Among competing hypotheses, the one with the fewest assumptions should be selected. Simpler theories are preferable because they are more amenable to testing.. I The principle of Occam’s razor is known in latin as lex parsimoniae — The law of parsimony — attributed to English philosopher (and Franciscan friar) William of Ockham (c.1287–1347).. I Ptolemy (c.AD90–c.AD168) stated an equivalent principle: “We consider it a good principle to explain the phenomena by the simplest hypothesis possible.”. I Isaac Newton (1642-1726), “We are to admit no more causes of natural things than such as are both true and sufficient to explain their appearances. Therefore, to the same natural effects we must, as far as possible, assign the same causes.” 24 / 84.

(50) Occam’s Razor Occam’s Razor Principle: Among competing hypotheses, the one with the fewest assumptions should be selected. Simpler theories are preferable because they are more amenable to testing.. I The principle of Occam’s razor is known in latin as lex parsimoniae — The law of parsimony — attributed to English philosopher (and Franciscan friar) William of Ockham (c.1287–1347).. I Ptolemy (c.AD90–c.AD168) stated an equivalent principle: “We consider it a good principle to explain the phenomena by the simplest hypothesis possible.”. I Isaac Newton (1642-1726), “We are to admit no more causes of natural things than such as are both true and sufficient to explain their appearances. Therefore, to the same natural effects we must, as far as possible, assign the same causes.” 24 / 84.

(51) Occam’s Razor Occam’s Razor Principle: Among competing hypotheses, the one with the fewest assumptions should be selected. Simpler theories are preferable because they are more amenable to testing.. I The principle of Occam’s razor is known in latin as lex parsimoniae — The law of parsimony — attributed to English philosopher (and Franciscan friar) William of Ockham (c.1287–1347).. I Ptolemy (c.AD90–c.AD168) stated an equivalent principle: “We consider it a good principle to explain the phenomena by the simplest hypothesis possible.”. I Isaac Newton (1642-1726), “We are to admit no more causes of natural things than such as are both true and sufficient to explain their appearances. Therefore, to the same natural effects we must, as far as possible, assign the same causes.” 24 / 84.

(52) Occam’s Razor. https://xkcd.com/609/ 25 / 84.

(53) All Models Are Wong Since all models are wrong, the scientist cannot obtain a “correct” one by excessive elaboration. On the contrary, following William of Occam, he should seek an economical description of natural phenomena. Just as the ability to devise simple but evocative models is the signature of the great scientist, so over-elaboration and over-parameterization is often the mark of mediocrity. George Box, Journal of the American Statistical Association, 1976.. 26 / 84.

(54) Research Ethics. 27 / 84.

(55) Research Ethics. http://dilbert.com/strip/2010-09-01/ Dilbert, September 1st, 2010.. 28 / 84.

(56) Research Ethics I Informed consent: When an individual understands the purpose and consents to being a subject of research.. I Honest cooperation is the main obligation of the research participant. In return, the research participant has the right to expect privacy and confidentiality.. I Subjects should be informed of their right to be left alone or to break off the interview at any time.. I Objective reporting: The researcher should not misrepresent results and should report difficulties, potential biases, limitations.. I The sponsor should respect participant confidentiality, not alter data to meet a desired objective, not introduce bias into interpretation of the study, not omit results, not make recommendations beyond the scope of the study.. 29 / 84.

(57) Research Ethics I Informed consent: When an individual understands the purpose and consents to being a subject of research.. I Honest cooperation is the main obligation of the research participant. In return, the research participant has the right to expect privacy and confidentiality.. I Subjects should be informed of their right to be left alone or to break off the interview at any time.. I Objective reporting: The researcher should not misrepresent results and should report difficulties, potential biases, limitations.. I The sponsor should respect participant confidentiality, not alter data to meet a desired objective, not introduce bias into interpretation of the study, not omit results, not make recommendations beyond the scope of the study.. 29 / 84.

(58) Research Ethics I Informed consent: When an individual understands the purpose and consents to being a subject of research.. I Honest cooperation is the main obligation of the research participant. In return, the research participant has the right to expect privacy and confidentiality.. I Subjects should be informed of their right to be left alone or to break off the interview at any time.. I Objective reporting: The researcher should not misrepresent results and should report difficulties, potential biases, limitations.. I The sponsor should respect participant confidentiality, not alter data to meet a desired objective, not introduce bias into interpretation of the study, not omit results, not make recommendations beyond the scope of the study.. 29 / 84.

(59) Research Ethics I Informed consent: When an individual understands the purpose and consents to being a subject of research.. I Honest cooperation is the main obligation of the research participant. In return, the research participant has the right to expect privacy and confidentiality.. I Subjects should be informed of their right to be left alone or to break off the interview at any time.. I Objective reporting: The researcher should not misrepresent results and should report difficulties, potential biases, limitations.. I The sponsor should respect participant confidentiality, not alter data to meet a desired objective, not introduce bias into interpretation of the study, not omit results, not make recommendations beyond the scope of the study.. 29 / 84.

(60) Research Ethics I Informed consent: When an individual understands the purpose and consents to being a subject of research.. I Honest cooperation is the main obligation of the research participant. In return, the research participant has the right to expect privacy and confidentiality.. I Subjects should be informed of their right to be left alone or to break off the interview at any time.. I Objective reporting: The researcher should not misrepresent results and should report difficulties, potential biases, limitations.. I The sponsor should respect participant confidentiality, not alter data to meet a desired objective, not introduce bias into interpretation of the study, not omit results, not make recommendations beyond the scope of the study.. 29 / 84.

(61) More and more organizations are accepting the fact that, despite their best efforts, security breaches are unavoidable.. BREACH LEVEL INDEX. THE NUMBERS. RECORDS BREACHED IN FIRST HALF OF 2017. 1,901,866,611 NUMBER OF BREACH INCIDENTS. 918. PERCENTAGE OF BREACHES WHERE NUMBER OF COMPROMISED RECORDS WAS UNKNOWN. 59.3% 4.6%. PERCENTAGE OF DATA BREACHES WHERE ENCRYPTION WAS USED. DATA RECORDS WERE LOST OR STOLEN WITH THE FOLLOWING FREQUENCY. Data Breach. EVERY DAY. 10,507,550 EVERY HOUR. 437,815 EVERY MINUTE. 7,297. EVERY SECOND. 122. http://breachlevelindex.com/. 30 / 84.

(62) Research Ethics I Leading questions: Designed to condition the subject about a particular issue, e.g. “Do you agree with killing innocent creatures? . . . Is it wrong to eat meat?”. I Loaded question: Suggests a socially desirable answer and/or is emotionally charged, e.g. “People have become more conscious about fuel emission. Would you consider buying an electric car?”. I Sponsored research: If the sponsor expects the research to support certain types of results, it is clearly unethical. Mixing sales or fund-raising with research is unethical. Examples: The gun lobby and the effects of guns on gun violence; the oil industry lobby and the effects of emissions on global warming.. I Pseudo-research: Research conducted not to gather information with a true research purpose, but rather to bolster an established point of view.. 31 / 84.

(63) Research Ethics I Leading questions: Designed to condition the subject about a particular issue, e.g. “Do you agree with killing innocent creatures? . . . Is it wrong to eat meat?”. I Loaded question: Suggests a socially desirable answer and/or is emotionally charged, e.g. “People have become more conscious about fuel emission. Would you consider buying an electric car?”. I Sponsored research: If the sponsor expects the research to support certain types of results, it is clearly unethical. Mixing sales or fund-raising with research is unethical. Examples: The gun lobby and the effects of guns on gun violence; the oil industry lobby and the effects of emissions on global warming.. I Pseudo-research: Research conducted not to gather information with a true research purpose, but rather to bolster an established point of view.. 31 / 84.

(64) Research Ethics I Leading questions: Designed to condition the subject about a particular issue, e.g. “Do you agree with killing innocent creatures? . . . Is it wrong to eat meat?”. I Loaded question: Suggests a socially desirable answer and/or is emotionally charged, e.g. “People have become more conscious about fuel emission. Would you consider buying an electric car?”. I Sponsored research: If the sponsor expects the research to support certain types of results, it is clearly unethical. Mixing sales or fund-raising with research is unethical. Examples: The gun lobby and the effects of guns on gun violence; the oil industry lobby and the effects of emissions on global warming.. I Pseudo-research: Research conducted not to gather information with a true research purpose, but rather to bolster an established point of view.. 31 / 84.

(65) Research Ethics I Leading questions: Designed to condition the subject about a particular issue, e.g. “Do you agree with killing innocent creatures? . . . Is it wrong to eat meat?”. I Loaded question: Suggests a socially desirable answer and/or is emotionally charged, e.g. “People have become more conscious about fuel emission. Would you consider buying an electric car?”. I Sponsored research: If the sponsor expects the research to support certain types of results, it is clearly unethical. Mixing sales or fund-raising with research is unethical. Examples: The gun lobby and the effects of guns on gun violence; the oil industry lobby and the effects of emissions on global warming.. I Pseudo-research: Research conducted not to gather information with a true research purpose, but rather to bolster an established point of view.. 31 / 84.

(66) Research Ethics: Leading Questions. Left: Sir Humphrey Appleby, Permanent Secretary for the Department of Administrative Affairs (Nigel Hawthorne, 1929-2001). Right: James George Hacker, Minister of Administrative Affairs (Paul Eddington, 1927-1995). Yes, Minister is a British sitcom written by Antony Jay and Jonathan Lynn aired by the BBC from 1980 to 1984. 32 / 84.

(67) Leading Questions: Yes Minister Sir Humphrey: You know what happens: nice young lady comes up to you. Obviously you want to create a good impression, you don’t want to look a fool, do you? Bernard Woolley: No Sir Humphrey: So she starts asking you some questions: Mr. Woolley, are you worried about the number of young people without jobs? Bernard Woolley: Yes Sir Humphrey: Are you worried about the rise in crime among teenagers? Bernard Woolley: Yes. Sir Humphrey: Do you think there is lack of discipline in our Comprehensive Schools? Bernard Woolley: Yes. Sir Humphrey: Do you think young people welcome some authority and leadership in their lives?. 33 / 84.

(68) Leading Questions: Yes Minister Bernard Woolley: Yes. Sir Humphrey: Do you think they respond to a challenge? Bernard Woolley: Yes. Sir Humphrey: Would you be in favour of reintroducing National Service? Bernard Woolley: Oh, well I suppose I might. Sir Humphrey: Yes or no? Bernard Woolley: Yes. Sir Humphrey: Of course you would, Bernard. After all you told you can’t say no to that. So they don’t mention the first five questions and they publish the last one. Bernard Woolley: Is that really what they do? Sir Humphrey: Well, not the reputable ones, no, but there aren’t many of those. So alternatively the young lady can get the opposite result. Bernard Woolley: How? Sir Humphrey: Mr. Woolley, are you worried about the danger of war?. 34 / 84.

(69) Leading Questions: Yes Minister Bernard Woolley: Yes. Sir Humphrey: Are you worried about the growth of armaments? Bernard Woolley: Yes. Sir Humphrey: Do you think there’s a danger in giving young people guns and teaching them how to kill? Bernard Woolley: Yes. Sir Humphrey: Do you think it’s wrong to force people to take arms against their will? Bernard Woolley: Yes. Sir Humphrey: Would you oppose the reintroduction of National Service? Bernard Woolley: Yes. Sir Humphrey: There you are, you see, Bernard. The perfect balanced sample.. 35 / 84.

(70) Research Ethics: Push Polls Push poll: Telemarketing under the disguise of research intended to promote a particular position or point of view. Would you still support Obama if Hamas had expressed support for him?. . . . but the Hamas leadership never expressed their support for Obama! And (obviously) neither did Obama for Hamas. Jonathan Cohn, “I Just Got Push-Polled on Obama and Israel,” The New Republic, September 15, 2008.. 36 / 84.

(71) Research Ethics: Push Polls Push poll: Telemarketing under the disguise of research intended to promote a particular position or point of view. Would you still support Obama if Hamas had expressed support for him?. . . . but the Hamas leadership never expressed their support for Obama! And (obviously) neither did Obama for Hamas. Jonathan Cohn, “I Just Got Push-Polled on Obama and Israel,” The New Republic, September 15, 2008.. 36 / 84.

(72) Survey Research. 37 / 84.

(73) Surveys. 38 / 84.

(74) Surveys I Survey: A survey is a method of collecting primary data based on communication with a sampling unit.. I Sampling units: A single element randomly selected from the population. Could be one person, a couple, a family.. I Respondent: Person who answers an interviewer’s questions. I Objectives: describe events or opinion about particular activities, e.g. Identify characteristics of target markets, describe employee job satisfaction, describe the risk aversion of financial investor. I Examples: Personal interviews, Telephone interviews, Mail surveys, Internet surveys, Self administrated questionnaires. I Margin of error: A measure of the amount of random sampling error in a survey’s results. Sampling theory provides methods for calculating the probability that the survey results differ from the sampled population by more than a certain amount, simply due to chance. The margin of error only accounts for random sampling error, ignoring other sources of errors. 39 / 84.

(75) Surveys I Survey: A survey is a method of collecting primary data based on communication with a sampling unit.. I Sampling units: A single element randomly selected from the population. Could be one person, a couple, a family.. I Respondent: Person who answers an interviewer’s questions. I Objectives: describe events or opinion about particular activities, e.g. Identify characteristics of target markets, describe employee job satisfaction, describe the risk aversion of financial investor. I Examples: Personal interviews, Telephone interviews, Mail surveys, Internet surveys, Self administrated questionnaires. I Margin of error: A measure of the amount of random sampling error in a survey’s results. Sampling theory provides methods for calculating the probability that the survey results differ from the sampled population by more than a certain amount, simply due to chance. The margin of error only accounts for random sampling error, ignoring other sources of errors. 39 / 84.

(76) Surveys I Survey: A survey is a method of collecting primary data based on communication with a sampling unit.. I Sampling units: A single element randomly selected from the population. Could be one person, a couple, a family.. I Respondent: Person who answers an interviewer’s questions. I Objectives: describe events or opinion about particular activities, e.g. Identify characteristics of target markets, describe employee job satisfaction, describe the risk aversion of financial investor. I Examples: Personal interviews, Telephone interviews, Mail surveys, Internet surveys, Self administrated questionnaires. I Margin of error: A measure of the amount of random sampling error in a survey’s results. Sampling theory provides methods for calculating the probability that the survey results differ from the sampled population by more than a certain amount, simply due to chance. The margin of error only accounts for random sampling error, ignoring other sources of errors. 39 / 84.

(77) Surveys I Survey: A survey is a method of collecting primary data based on communication with a sampling unit.. I Sampling units: A single element randomly selected from the population. Could be one person, a couple, a family.. I Respondent: Person who answers an interviewer’s questions. I Objectives: describe events or opinion about particular activities, e.g. Identify characteristics of target markets, describe employee job satisfaction, describe the risk aversion of financial investor. I Examples: Personal interviews, Telephone interviews, Mail surveys, Internet surveys, Self administrated questionnaires. I Margin of error: A measure of the amount of random sampling error in a survey’s results. Sampling theory provides methods for calculating the probability that the survey results differ from the sampled population by more than a certain amount, simply due to chance. The margin of error only accounts for random sampling error, ignoring other sources of errors. 39 / 84.

(78) Surveys I Survey: A survey is a method of collecting primary data based on communication with a sampling unit.. I Sampling units: A single element randomly selected from the population. Could be one person, a couple, a family.. I Respondent: Person who answers an interviewer’s questions. I Objectives: describe events or opinion about particular activities, e.g. Identify characteristics of target markets, describe employee job satisfaction, describe the risk aversion of financial investor. I Examples: Personal interviews, Telephone interviews, Mail surveys, Internet surveys, Self administrated questionnaires. I Margin of error: A measure of the amount of random sampling error in a survey’s results. Sampling theory provides methods for calculating the probability that the survey results differ from the sampled population by more than a certain amount, simply due to chance. The margin of error only accounts for random sampling error, ignoring other sources of errors. 39 / 84.

(79) Surveys I Survey: A survey is a method of collecting primary data based on communication with a sampling unit.. I Sampling units: A single element randomly selected from the population. Could be one person, a couple, a family.. I Respondent: Person who answers an interviewer’s questions. I Objectives: describe events or opinion about particular activities, e.g. Identify characteristics of target markets, describe employee job satisfaction, describe the risk aversion of financial investor. I Examples: Personal interviews, Telephone interviews, Mail surveys, Internet surveys, Self administrated questionnaires. I Margin of error: A measure of the amount of random sampling error in a survey’s results. Sampling theory provides methods for calculating the probability that the survey results differ from the sampled population by more than a certain amount, simply due to chance. The margin of error only accounts for random sampling error, ignoring other sources of errors. 39 / 84.

(80) Margin of Error On almost every occasion when we release a new survey, someone in the media will ask, “What is the margin of error for this survey?” There is only one honest and accurate answer to this question — which I sometimes use to the great confusion of my audience — and that is, “The possible margin of error is infinite.” Humphrey Taylor, Chairman of the Harris Poll.. 40 / 84.

(81) Margin of Error 2016 U.S. Presidential Election Poll Results. 41 / 84.

(82) Errors in Surveys I Random Sampling Error: Occurs because of chance variation in the sampling units selected for a particular sample.. I Systematic Error: Occurs because of some imperfect aspect of the research design or from persistent mistakes in execution.. I Sample Bias: Occurs when samples deviate systematically from the (unobserved) population. May be caused by flawed sample design or improper sampling procedure execution. I Self-Selection Bias: Occurs because respondents who feel strongly about a subject are more likely to respond to survey questions than people who feel indifferent about it. I Response Bias: Occurs when respondents either consciously or unconsciously tend to answer questions with a certain slant that misrepresents the truth. I Nonresponse Error: Occurs when the sampling units selected for a sample are not interviewed, because they are unavailable or unwilling to take part. 42 / 84.

(83) Errors in Surveys I Random Sampling Error: Occurs because of chance variation in the sampling units selected for a particular sample.. I Systematic Error: Occurs because of some imperfect aspect of the research design or from persistent mistakes in execution.. I Sample Bias: Occurs when samples deviate systematically from the (unobserved) population. May be caused by flawed sample design or improper sampling procedure execution. I Self-Selection Bias: Occurs because respondents who feel strongly about a subject are more likely to respond to survey questions than people who feel indifferent about it. I Response Bias: Occurs when respondents either consciously or unconsciously tend to answer questions with a certain slant that misrepresents the truth. I Nonresponse Error: Occurs when the sampling units selected for a sample are not interviewed, because they are unavailable or unwilling to take part. 42 / 84.

(84) Errors in Surveys I Random Sampling Error: Occurs because of chance variation in the sampling units selected for a particular sample.. I Systematic Error: Occurs because of some imperfect aspect of the research design or from persistent mistakes in execution.. I Sample Bias: Occurs when samples deviate systematically from the (unobserved) population. May be caused by flawed sample design or improper sampling procedure execution. I Self-Selection Bias: Occurs because respondents who feel strongly about a subject are more likely to respond to survey questions than people who feel indifferent about it. I Response Bias: Occurs when respondents either consciously or unconsciously tend to answer questions with a certain slant that misrepresents the truth. I Nonresponse Error: Occurs when the sampling units selected for a sample are not interviewed, because they are unavailable or unwilling to take part. 42 / 84.

(85) Errors in Surveys I Random Sampling Error: Occurs because of chance variation in the sampling units selected for a particular sample.. I Systematic Error: Occurs because of some imperfect aspect of the research design or from persistent mistakes in execution.. I Sample Bias: Occurs when samples deviate systematically from the (unobserved) population. May be caused by flawed sample design or improper sampling procedure execution. I Self-Selection Bias: Occurs because respondents who feel strongly about a subject are more likely to respond to survey questions than people who feel indifferent about it. I Response Bias: Occurs when respondents either consciously or unconsciously tend to answer questions with a certain slant that misrepresents the truth. I Nonresponse Error: Occurs when the sampling units selected for a sample are not interviewed, because they are unavailable or unwilling to take part. 42 / 84.

(86) Errors in Surveys I Random Sampling Error: Occurs because of chance variation in the sampling units selected for a particular sample.. I Systematic Error: Occurs because of some imperfect aspect of the research design or from persistent mistakes in execution.. I Sample Bias: Occurs when samples deviate systematically from the (unobserved) population. May be caused by flawed sample design or improper sampling procedure execution. I Self-Selection Bias: Occurs because respondents who feel strongly about a subject are more likely to respond to survey questions than people who feel indifferent about it. I Response Bias: Occurs when respondents either consciously or unconsciously tend to answer questions with a certain slant that misrepresents the truth. I Nonresponse Error: Occurs when the sampling units selected for a sample are not interviewed, because they are unavailable or unwilling to take part. 42 / 84.

(87) Errors in Surveys I Random Sampling Error: Occurs because of chance variation in the sampling units selected for a particular sample.. I Systematic Error: Occurs because of some imperfect aspect of the research design or from persistent mistakes in execution.. I Sample Bias: Occurs when samples deviate systematically from the (unobserved) population. May be caused by flawed sample design or improper sampling procedure execution. I Self-Selection Bias: Occurs because respondents who feel strongly about a subject are more likely to respond to survey questions than people who feel indifferent about it. I Response Bias: Occurs when respondents either consciously or unconsciously tend to answer questions with a certain slant that misrepresents the truth. I Nonresponse Error: Occurs when the sampling units selected for a sample are not interviewed, because they are unavailable or unwilling to take part. 42 / 84.

(88) Errors in Surveys: Literary Digest, 1936 The Literary Digest had correctly predicted the outcomes of the 1916, 1920, 1924, 1928, and 1932 elections by conducting polls. In 1936, based on “straw” ballots, they predicted Alf Landon would beat Franklin Delanor Roosevelt 57% to 43% — Roosevelt won 61% to 37%, based on 46 million votes, winning in 46/48 states (98.49% of the electoral vote). The sample was biased: 10 million ballots had been mailed, only 2.4 million returned — a 75% non-response rate — using lists of phone numbers, drivers’ registrations, and country club memberships. This was the middle of the depression: voters who supported Roosevelt were less likely to respond to the survey.. The Literary Digest folded soon after this historic blunder.. 43 / 84.

(89) Errors in Surveys: Literary Digest, 1936 Nation-wide county map shaded by voter share, U.S. presidential election 1936.. https://en.wikipedia.org/wiki/United_States_presidential_election,_1936. 44 / 84.

(90) Errors in Surveys: Non-Response Bias. http://www.francartoons.co.uk. 45 / 84.

(91) Errors in Surveys: Respondents I Unconscious misrepresentation: When a respondent is consciously trying to be truthful and cooperative, response bias can arise from the question format, the question content, or some other stimulus that affects their response to a question.. I Acquiescence Bias: A tendency to agree with most questions. I Extremity Bias: A tendency to select extreme responses. I Auspices Bias: Occurs when answers reflect the person’s liking or disliking of the sponsor organization.. I Social Desirability Bias: Occurs when respondents’ answers are biased by their desire, either conscious or unconscious, to be viewed favourably, e.g. over-reporting “good behavior”.. I Deliberate falsification: can sometimes be detected.. 46 / 84.

(92) Errors in Surveys: Respondents I Unconscious misrepresentation: When a respondent is consciously trying to be truthful and cooperative, response bias can arise from the question format, the question content, or some other stimulus that affects their response to a question.. I Acquiescence Bias: A tendency to agree with most questions. I Extremity Bias: A tendency to select extreme responses. I Auspices Bias: Occurs when answers reflect the person’s liking or disliking of the sponsor organization.. I Social Desirability Bias: Occurs when respondents’ answers are biased by their desire, either conscious or unconscious, to be viewed favourably, e.g. over-reporting “good behavior”.. I Deliberate falsification: can sometimes be detected.. 46 / 84.

(93) Errors in Surveys: Respondents I Unconscious misrepresentation: When a respondent is consciously trying to be truthful and cooperative, response bias can arise from the question format, the question content, or some other stimulus that affects their response to a question.. I Acquiescence Bias: A tendency to agree with most questions. I Extremity Bias: A tendency to select extreme responses. I Auspices Bias: Occurs when answers reflect the person’s liking or disliking of the sponsor organization.. I Social Desirability Bias: Occurs when respondents’ answers are biased by their desire, either conscious or unconscious, to be viewed favourably, e.g. over-reporting “good behavior”.. I Deliberate falsification: can sometimes be detected.. 46 / 84.

(94) Errors in Surveys: Respondents I Unconscious misrepresentation: When a respondent is consciously trying to be truthful and cooperative, response bias can arise from the question format, the question content, or some other stimulus that affects their response to a question.. I Acquiescence Bias: A tendency to agree with most questions. I Extremity Bias: A tendency to select extreme responses. I Auspices Bias: Occurs when answers reflect the person’s liking or disliking of the sponsor organization.. I Social Desirability Bias: Occurs when respondents’ answers are biased by their desire, either conscious or unconscious, to be viewed favourably, e.g. over-reporting “good behavior”.. I Deliberate falsification: can sometimes be detected.. 46 / 84.

(95) Errors in Surveys: Respondents I Unconscious misrepresentation: When a respondent is consciously trying to be truthful and cooperative, response bias can arise from the question format, the question content, or some other stimulus that affects their response to a question.. I Acquiescence Bias: A tendency to agree with most questions. I Extremity Bias: A tendency to select extreme responses. I Auspices Bias: Occurs when answers reflect the person’s liking or disliking of the sponsor organization.. I Social Desirability Bias: Occurs when respondents’ answers are biased by their desire, either conscious or unconscious, to be viewed favourably, e.g. over-reporting “good behavior”.. I Deliberate falsification: can sometimes be detected.. 46 / 84.

(96) Errors in Surveys: Respondents I Unconscious misrepresentation: When a respondent is consciously trying to be truthful and cooperative, response bias can arise from the question format, the question content, or some other stimulus that affects their response to a question.. I Acquiescence Bias: A tendency to agree with most questions. I Extremity Bias: A tendency to select extreme responses. I Auspices Bias: Occurs when answers reflect the person’s liking or disliking of the sponsor organization.. I Social Desirability Bias: Occurs when respondents’ answers are biased by their desire, either conscious or unconscious, to be viewed favourably, e.g. over-reporting “good behavior”.. I Deliberate falsification: can sometimes be detected.. 46 / 84.

(97) Errors in Surveys: Dishonesty. How We Lie to Everyone — Especially Ourselves. 47 / 84.

(98) Errors in Surveys: Deliberate falsification. http://www.gocomics.com/calvinandhobbes/1995/08/23 Calvin and Hobbes, August 23rd, 1995.. 48 / 84.

(99) Errors in Surveys: Researchers I Administrative error: An error caused by the improper administration or execution of the research task.. I Data-Processing Error: incorrect data entry, incorrect computer programming, or other procedural errors during data analysis.. I Interviewer Error: mistakes made by interviewers, failing to record survey responses correctly.. I Interviewer Cheating: filling in fake answers or falsifying questionnaires by an interviewer.. 49 / 84.

(100) Errors in Surveys: Researchers I Administrative error: An error caused by the improper administration or execution of the research task.. I Data-Processing Error: incorrect data entry, incorrect computer programming, or other procedural errors during data analysis.. I Interviewer Error: mistakes made by interviewers, failing to record survey responses correctly.. I Interviewer Cheating: filling in fake answers or falsifying questionnaires by an interviewer.. 49 / 84.

(101) Errors in Surveys: Researchers I Administrative error: An error caused by the improper administration or execution of the research task.. I Data-Processing Error: incorrect data entry, incorrect computer programming, or other procedural errors during data analysis.. I Interviewer Error: mistakes made by interviewers, failing to record survey responses correctly.. I Interviewer Cheating: filling in fake answers or falsifying questionnaires by an interviewer.. 49 / 84.

(102) Errors in Surveys: Researchers I Administrative error: An error caused by the improper administration or execution of the research task.. I Data-Processing Error: incorrect data entry, incorrect computer programming, or other procedural errors during data analysis.. I Interviewer Error: mistakes made by interviewers, failing to record survey responses correctly.. I Interviewer Cheating: filling in fake answers or falsifying questionnaires by an interviewer.. 49 / 84.

(103) Errors in Surveys: Interviewer Cheating. http://freakonomics.com/ Schoolteachers caught cheating. 50 / 84.

(104) Questionnaire Design. 51 / 84.

(105) https://xkcd.com/1289/ 52 / 84.

(106) Questionnaires I In a questionnaire, the questions must meet the basic criteria of relevance and accuracy.. I Questionnaire Relevancy: Collected information should address the research question.. I The researcher should be specific about data needs and have rationale for each item requesting information.. I Irrelevant questions will make the survey needlessly long. I Questionnaire Accuracy: Obtaining accurate answers from respondents depend on the questionnaire design, to facilitate recall and motivate respondents to cooperate.. 53 / 84.

(107) Questionnaires I In a questionnaire, the questions must meet the basic criteria of relevance and accuracy.. I Questionnaire Relevancy: Collected information should address the research question.. I The researcher should be specific about data needs and have rationale for each item requesting information.. I Irrelevant questions will make the survey needlessly long. I Questionnaire Accuracy: Obtaining accurate answers from respondents depend on the questionnaire design, to facilitate recall and motivate respondents to cooperate.. 53 / 84.

(108) Questionnaires I In a questionnaire, the questions must meet the basic criteria of relevance and accuracy.. I Questionnaire Relevancy: Collected information should address the research question.. I The researcher should be specific about data needs and have rationale for each item requesting information.. I Irrelevant questions will make the survey needlessly long. I Questionnaire Accuracy: Obtaining accurate answers from respondents depend on the questionnaire design, to facilitate recall and motivate respondents to cooperate.. 53 / 84.

(109) Questionnaires I In a questionnaire, the questions must meet the basic criteria of relevance and accuracy.. I Questionnaire Relevancy: Collected information should address the research question.. I The researcher should be specific about data needs and have rationale for each item requesting information.. I Irrelevant questions will make the survey needlessly long. I Questionnaire Accuracy: Obtaining accurate answers from respondents depend on the questionnaire design, to facilitate recall and motivate respondents to cooperate.. 53 / 84.

(110) Questionnaires I In a questionnaire, the questions must meet the basic criteria of relevance and accuracy.. I Questionnaire Relevancy: Collected information should address the research question.. I The researcher should be specific about data needs and have rationale for each item requesting information.. I Irrelevant questions will make the survey needlessly long. I Questionnaire Accuracy: Obtaining accurate answers from respondents depend on the questionnaire design, to facilitate recall and motivate respondents to cooperate.. 53 / 84.

(111) Questionnaires I Open-Ended Questions: Pose some problem and ask respondents to answer in their own words. Example: “What things do you like most about studying in Hong Kong Baptist University?” Are most beneficial in exploratory research, especially when the range of responses is not known. High cost of administering open-ended response questions. The analysis is subjected to the interpretation of the researcher and may reflect a combination of the respondent’s and interviewer’s ideas.. 54 / 84.

(112) Questionnaires I Open-Ended Questions: Pose some problem and ask respondents to answer in their own words. Example: “What things do you like most about studying in Hong Kong Baptist University?” Are most beneficial in exploratory research, especially when the range of responses is not known. High cost of administering open-ended response questions. The analysis is subjected to the interpretation of the researcher and may reflect a combination of the respondent’s and interviewer’s ideas.. 54 / 84.

(113) Questionnaires I Open-Ended Questions: Pose some problem and ask respondents to answer in their own words. Example: “What things do you like most about studying in Hong Kong Baptist University?” Are most beneficial in exploratory research, especially when the range of responses is not known. High cost of administering open-ended response questions. The analysis is subjected to the interpretation of the researcher and may reflect a combination of the respondent’s and interviewer’s ideas.. 54 / 84.

(114) Questionnaires I Open-Ended Questions: Pose some problem and ask respondents to answer in their own words. Example: “What things do you like most about studying in Hong Kong Baptist University?” Are most beneficial in exploratory research, especially when the range of responses is not known. High cost of administering open-ended response questions. The analysis is subjected to the interpretation of the researcher and may reflect a combination of the respondent’s and interviewer’s ideas.. 54 / 84.

(115) Questionnaires I Open-Ended Questions: Pose some problem and ask respondents to answer in their own words. Example: “What things do you like most about studying in Hong Kong Baptist University?” Are most beneficial in exploratory research, especially when the range of responses is not known. High cost of administering open-ended response questions. The analysis is subjected to the interpretation of the researcher and may reflect a combination of the respondent’s and interviewer’s ideas.. 54 / 84.

(116) Questionnaires I Fixed-Alternate Questions: Questions in which respondents are given specific, limited-alternative responses and asked to choose the one closest to their own. Researcher needs to know the potential responses to a question. The alternative should be mutually exclusive and collectively exhaustive. Example: Would you say the labor quality in Hong Kong is higher, about the same, or not as good as it was 10 years ago? Higher / About the same / Not as good. Require less interviewer skill, less time, is easier for the respondent to answer, and easier to analyze.. 55 / 84.

(117) Questionnaires I Fixed-Alternate Questions: Questions in which respondents are given specific, limited-alternative responses and asked to choose the one closest to their own. Researcher needs to know the potential responses to a question. The alternative should be mutually exclusive and collectively exhaustive. Example: Would you say the labor quality in Hong Kong is higher, about the same, or not as good as it was 10 years ago? Higher / About the same / Not as good. Require less interviewer skill, less time, is easier for the respondent to answer, and easier to analyze.. 55 / 84.

(118) Questionnaires I Fixed-Alternate Questions: Questions in which respondents are given specific, limited-alternative responses and asked to choose the one closest to their own. Researcher needs to know the potential responses to a question. The alternative should be mutually exclusive and collectively exhaustive. Example: Would you say the labor quality in Hong Kong is higher, about the same, or not as good as it was 10 years ago? Higher / About the same / Not as good. Require less interviewer skill, less time, is easier for the respondent to answer, and easier to analyze.. 55 / 84.

(119) Questionnaires I Fixed-Alternate Questions: Questions in which respondents are given specific, limited-alternative responses and asked to choose the one closest to their own. Researcher needs to know the potential responses to a question. The alternative should be mutually exclusive and collectively exhaustive. Example: Would you say the labor quality in Hong Kong is higher, about the same, or not as good as it was 10 years ago? Higher / About the same / Not as good. Require less interviewer skill, less time, is easier for the respondent to answer, and easier to analyze.. 55 / 84.

(120) Questionnaires I Fixed-Alternate Questions: Questions in which respondents are given specific, limited-alternative responses and asked to choose the one closest to their own. Researcher needs to know the potential responses to a question. The alternative should be mutually exclusive and collectively exhaustive. Example: Would you say the labor quality in Hong Kong is higher, about the same, or not as good as it was 10 years ago? Higher / About the same / Not as good. Require less interviewer skill, less time, is easier for the respondent to answer, and easier to analyze.. 55 / 84.

References

Related documents