Study Guide Final Exam Quantitative Research research that provides data that can be expressed with numbers such as ranks or scales o Goal precision o Measurement Accurate information accurate decision making Better chances of being successful Operationalizing specifies what the researcher must do to measure the concept under investigation Scale series of items arranged according to value for the purpose of quantification Need at least 3 items Levels of Measurement Nominal categorizes labeled data it either is or isn t Specifies gender indicate yes or no check all that apply o General attributes are only named weakest o Nominal Level Questions Appropriate for categorizing responses Useful when the main purpose is to describe patterns among the sample Multiple choice checklist Options should avoid response bias Ordinal rank orders data without specifying magnitude of distance NOT NUMBERS o Rank from most important to least important 1 10 o RANK attributes can be ordered o Ordinal Scale Properties Uniquely classifies while preserving order important when order of response is important Interval rank order with comparable magnitude between choices assumed equal o Indicate your level of agreement meaningful number Ex Between how much money do you spend a month o Specific distance is meaningful o Interval Scale Properties Uniquely classifies preserves order equal intervals Appropriate when responses to an explicit well defined question is needed o Types of Interval Rating scales bipolar Staple scale Semantic differential adjective anchors simplified agreement disagreement scales Ratio N O I characteristics with verifiable Likert scale semantic differential absolute distances between choices o Allocate a bonus of 10 000 between you and your team members o Absolute Zero o Ratio Scale Properties Appropriate when degree of behavior attitude or relative weight is important Uniquely classifies preserves order and has equal intervals natural zero weight and distance o Sampling Techniques Random sampling Quota sampling Snowball sampling o Questionnaire the specific set of questions that respondents answer A good question flows logically and has questions that are clear and meaningful to the respondent Gives solid instructions that guide the respondent through Design Page 324 in book Avoid blasting questions asked later in the survey by earlier questions Proceed within any certain topic from the most general questions to the most specific Reflect accepted industry procedures with regard to question sequence Be complete o Surveys a series of formatted questions delivered to a defined sample of people respondents Why Surveys Types of Surveys snapshot Quick inexpensive efficient accurate and flexible Cross sectional one time survey like a Trend different groups of similar groups Panel Studies follow same people over time Cohort specific group such as target audience Longitudinal describes research that measures a trait in a particular group of subjects over a long period of time o Non response Error People who refuse to cooperate they must answer all the questions Not at homes bias Only sample people at home creates a sample Similar to online surveys only get people with internet access or that visit a particular location Most non response error comes from respondents not answering individual questions while taking the survey Respondents skip questions for many reasons We will not know why less precision and accuracy o Biases Response Bias Extremity Bias A bias that occurs when respondents tend to answer questions with a certain slant that consciously or unconsciously misrepresents the truth A category of response bias that results because response styles vary from person to person Some individuals tend to use extremes when responding to questions o This is even more powerful among westerners those from a Judeo Christian culture who believe they will have to justify their answers Auspicious Bias Results because some individuals tend to agree with all questions or to concur with a particular position Bias in the responses of subjects caused by the respondents being influenced by the organization conducting the study Be aware of social desirability even in blind or anonymous samples Social Desirability Bias When respondents feel compelled to answer a specific way based on the environment The desire to answer and gain prestige or appear in a different social Acquiescence Bias Results because more individual tend to agree with all questions or to concur with a particular position Ex YES YES YES Interviewer Bias Occurs because the presence of the interviewer influences answers o Administrative Error Improper administration of the research task incorrect data entry filling in fake answers or Data processing error Interview cheating falsifying interviews computer programming or other procedural errors during the analysis stage Sample selection error or sampling procedure execution field mistake Interviewer error improper sample design Blunders Confusion Neglect Omission Qualitative data should be analyzed by someone unfamiliar with the study hypothesis Open Closed ended o Closed pre defined response categories 4 Levels Nominal Ordinal Interval Ratio Provides the answer o Open encourages free response Open Ended questions are analyzed as Nominal data Charts o Pie Chart Best for seeing how something is divided among different groups Shows categories of data in portion to the whole set Ex What types of books do you enjoy Check all that Shows as 100 then what they chose divided up apply o Bar graph Useful for comparing categories of data where you can have a single category of data OR You can break down the data further into multiple categories for greater depth of analysis What makes a good questionnaire o Introduction a statement that explains the purpose and goals of the research and asks for the respondent s cooperation and participation in the research study Common elements include An explicit or implicit reference to importance General information on the rationale and goals of the research An explicit request for participation Reassurance that the task of participating is not too burdensome or time consuming The need for truthful answers The promise of confidentiality The reassurance that is real legitimate research o Screener admits into the study all individuals who possess all of the target characteristics and eliminates all individuals who lack at least one of the identified characteristics