business researce final exame
Measures Produced in EDA
There are a range of descriptive measures which are produced on data using the EDA facility. These include:
n A range of Descriptive Statistics (mean, median, standard deviation, ranges, skewness)
n Confidence Intervals
n Stem-and-Leaf plots
n Boxplots
n Histograms (can be complemented with other plots, such as bar charts)
Analyses Available
Ratio and Interval data, from questionnaires, surveys or observations can be described and explored by the EDA approach, descriptive statistics are produced. However with knowledge of the sample size, mean, standard deviation and providing a representative sample has been obtained, some inferential statistics can also be undertaken.
Example
What follows is a set of descriptive measures undertaken on the variable Age, which is ratio data, using the explore facility of SPSS:
Dependence on Others
Are any of the variables dependent on one or more of the others? - If we are to assume that one or more of our variables may be explained and consequently predicted by at least two other independent variables then dependence is involved.
Number Dependent
Is there more than one dependent variable?
This can also be answered by the first question. If the answer is no then techniques such as multiple regression are suitable. If it is yes then this technique, amongst others, is not suitable, and specialised areas such as multivariate analysis of variance would need to be employed.
Nature of Data
What is the nature of the data? The scale of measurement has to be considered, that is: is it nominal or ordinal scale (metric), or is it interval or ratio scale (non-metric). The subsequent analyses of primary and secondary data, using multivariate techniques has rules which govern the sequence and approach to analysis. These are best summarised in the classification diagram (see earlier).
Methods of Analysis
Analysis for the different techniques can be facilitated on SPSS. You will look at approaches and printouts of the following techniques:
n Dependence Techniques - One dependent variable and multiple dependent variables
n Interdependence Techniques - Focus on variables and focus on objects
Dependence Techniques
n One dependent variable include Multiple Regression, Discriminant Analysis and Conjoint Analysis
n Multiple dependent variables deals with Canonical Correlation
Multiple Regression
This examines the relationship between at least two interval scaled independent variables and one interval scaled dependent variable. It differs from regression analysis, which is bivariate (involving one dependent and one independent variable).
Here means are positioned from smallest to largest and the distance or number of steps that two means are apart in this ranking, are used, in computing the range value for each comparison.
The test is based on the assumption that the larger the number of means being compared, the more likely that significantly different comparisons will occur.
Multiple Range Test
Canonical Correlation
This is an extension of simple correlation analysis, to the situations involving two or more independent variables and their degrees of association with the independent variable. There are two coefficients used: The coefficient of multiple correlation, R which indicates the strength of relationship between two or more independent variables and the dependent variable. Also the coefficient of multiple determination, R2
it indicates the proportion of variance in the dependent variable which is statistically accounted for by knowledge of the two (or more) independent variables. (see multiple regression)
5 Grounded theories
Grounded Theory is most accurately described as a research method in which the theory is developed from the data, rather than the other way around. That makes this is an inductive approach, meaning that it moves from the specific to the more general. It is mainly used for qualitative research, but is also applicable to other data (e.g., quantitative data;
7 Inferential statistics: If a Researcher uses data gathered on a sample and uses the statistics generated to reach conclusions about the population from which the sample was taken the statistics are inferential statistics. The two major types of inferential statistics are parametric statistics and non-parametric statistics. These types of data give rise to analysis divisions for data types
The Double Hermeneutic
The work of Giddens is significant here for in a number of works he further develops this form of abductive reasoning. By suggesting that the relationship between social science concepts and lay or everyday concepts is two way, social actors can re-appropriate social science concepts into everyday language and life.
Think for instance of the concept of social class.
(Giddens, 1982, Profiles and Critiques in Social Theory)
10 Philosophical concepts: Epistemology and Ontology are two of the most central concepts in the philosophy of science and social research.
Epistemology Is a theory of Knowledge, it is interested in the origins and the nature of knowing, and the construction of knowledge, it refers to the claims and assumptions that are made about the ways in which it is possible to gain knowledge of reality, about how what exists can be known. It is the inquiry into the conditions of the possibility of knowledge.
Informed Consent
The principle of informed consent rests on the idea that human research subjects should be able to agree to participate, or to not participate in research in the light of comprehensive information about the nature and purpose of the research. It is based on the assumption that individuals have a right to know what is happening to them.
(Homan, 1991)
Problems with Informed Consent
n The right to not know
n Based on assumption that individuals have the right to refuse - what about the Census
n Assumes that individuals know what they are disclosing
n Deployed as an exercise in persuasion - often a
tension between the researchers desire to get a good response rate, and individuals right to refuse
n Used as a justification device by researchers - so provides more protection for the researcher than the researched
n What about secondary analysis of the data
Posted by seangkhun
on 5:24 AM. Filed under
CASE STUDY
.
You can follow any responses to this entry through the RSS 2.0.
Feel free to leave a response