BUSI820 DB1 - Variables, Research Questions, and Data Coding PDF

Preview

Summary

Variables, Research Questions, and Data Coding...

Description

Discussion Board Forum 1 – Variables, Research Questions, and Data Coding

BUSI820 – Quantitative Research Methods Liberty University D1.1 Variables Independent and independent variables, one variable is usually the effect or cause of another. The purpose of research studies is to find out unknown characteristics or qualities of something or a person. For the qualities to get rated, the variables need to get defined. An independent variable usually influences the dependent variable directly or indirectly. Independent variables are of two types, which are active and attributes. An active independent variable can get its values manipulated to study how it affects other variables; therefore, it is necessary to infer cause (Morgan et al., 2013). However, one cannot always infer cause with the active variable as not all aspects can get changed. For example, anxiety levels can get altered to determine whether pain reduction medication affects patients' responsiveness. On the other hand, causal inferences are questionable when attribute variables cannot get changed, such as a person's age or nationality, but people of different ages can get studied. D1.2 Research Questions I Researchers use research questions to give answers, solutions, or insight on current world problems. There are three different types of questions which are different, associational, and descriptive. Descriptive research questions are aimed at given details concerning the variables to get measured (Morgan et al., 2013). The descriptive question aims to quantify the variables getting studied and uses words such as what percentage? What are? What is? How much?

Among others. On the other hand, the difference question is also known as comparative research questions. These questions aim to scrutinize the difference between one or more groups and the dependent variable. The questions start by asking what the difference is between? A particular dependent variable and a certain group of people. Lastly, associational questions are also referred to as relationship research questions. These questions are usually interested in trends, interactions, associations, or casual relationships among several group variables. Most associational questions are phrased, such as what is the relationship between? Certain independent variables and dependent variables. D1.3 Research Questions II D1.3a Association Question Examples of the research question will be bases on HSB variables such as mosaic pattern, visualization score, and religion. Example of an association question include: 1. What is the relationship between mosaic pattern and wealth in ancient Rome? 2. What is the relationship between creativity and data visualization among technology firms? 3. What is the relationship between faith and reason in religion? D1.3b Difference Question Example of the different questions include: 1. What is the difference between the mosaic pattern and the various types of patterns in art? 2. What is the difference between creative visualization and data visualization? 3. What is the difference between an atheist, a pagan, and a believer?

D1.3c Descriptive Question Examples of descriptive questions include: 1. How often do people in America opt for Mosaic patterns for decorating their houses? 2. What proportion of banks implements visualization scores in their business operation in the US? 3. How often do religious people fall into temptation after they have decided to follow the righteous way? D1.4 Data Coding I I think several changes need to add for the data coding of questionnaires to make them more reliable. For instance, the questionnaire which connects the researcher, interviewer, and respondent should be readable and easily analyzed by the respondent. The questionnaire should not become too hard to understand as the respondent might get bored. On the other hand, the computer will also find it difficult when coding the responses in a much understandable format. More so, it should be concise as it takes a lot of time to grab someone's attention and seconds to lose it. The questions should be complete and straight to the point, which will make coding easier. The researcher should write up a list of topics that they want to get information from to avoid asking questions that are unrelated to the survey. Another change is to ensure that the language used on the questionnaire and coding of data is not too technical for easier understanding. D1.5 Data Coding II The completed questionnaire in chapter two, problem 2.1, had several problems. The questionnaire suggested that the respondents circle or supply their answers. The instructions were

not quite clear at the very end; hence there is no uniformity, especially from participant seven to twelve. The lack of uniformity will make it difficult when coding the data to come up with overall easy-to-understand information (Morgan et al., 2013). For example, some participants choose two answers; others opted to indicate an X or circle, while others decided to write down. The questionnaire by participant ten was incomplete therefore cannot be used to compute the final data. Pilot tests would best suit the problem where the researcher gets to ask participants about the clarity of the items and what needs to be added or removed. Another solution would be to involve experts to evaluate the content validity to ensure all aspects are covered. When the questionnaires are not completed as desired, the researcher should then work on the completed ones for valid results. D1.6 Data Coding III It is important to check raw data before entering it into the computer for coding to ensure that all the sheets are marked appropriately. Checking before enables the researcher to find out which of the questionnaires have information relevant to the study and are complete with all the items correctly marked. More so, when checking, the researcher will get to see whether there are double answers for items that require a single answer and consider the particularly marked sheet not relevant to the study. Additionally, checking raw data is essential to ensure that the researcher can clean up the collected data and ensure it is consistent, readable, and clear before coding begins (Scherbaum & Shockley, 2015). On the other hand, errors that occur in coding or during data entry can also get solved. For instance, an entry outside the range of codes allocated to a particular variable will show on the table. This problem can get solved by the questionnaires were numbers and will require checking the respondent's number and finding out where the wrong codes have been entered.

References Morgan, G., Leech, N., Gloeckner, G., Barrett, K. (2013). IBM SPSS for Introductory Statistics (5th Ed.). New York, NY Scherbaum, C., & Shockley, K. (2015). Analyzing quantitative data for business and management students. Sage....