September, 2014 - PhD Thesis Bangalore

How do I Choose a Statistical Software

The world around is talking about data analysis. Whether one talks about analysis of consumer behaviour or a perspective to the critical metrics of six sigma or maybe any other programme that is actually data driven it is all related to data analysis. The good news here is that not only is the available data … Continue reading “How do I Choose a Statistical Software”

The world around is talking about data analysis. Whether one talks about analysis of consumer behaviour or a perspective to the critical metrics of six sigma or maybe any other programme that is actually data driven it is all related to data analysis. The good news here is that not only is the available data more than ever before but there are available an enormous range of software options which simplify the understanding of the data which otherwise is difficult to comprehend.

The options that are available for data analysis have a large gamut of options available. They range and vary from paper pencil option to a calculator to a more customized system that would very precisely take care of the smallest of detail tailored as per the needs of the researcher. However they would cost much more than the conventional systems and could go up to millions of rupees.

Talking of the extreme ends of the gamut, unless the researcher really enjoys calculations on fingers or has a huge amount of money lying idol to splurge, a software package sitting somewhere between these two extremes is what would work the best. However, that still leaves the researcher a wide variety of software packages to choose from and one needs to administer in a research.

When one talks about picking up data analysis software, there isn’t any right or wrong choice. What works best for a particular researcher may depend on more than one factor.

The first factor to consider is to analyse the person who will be using the software. The statistical skills of the person in terms of, whether he is an expert, novice or a blend of both. Will the data be analysed day in day out or once in a while? When this is figured out, it helps to match options with needs so as it can be avoided to choose any wrong package that is either too difficult to handle or does the wrong thing entirely

Comparison of Data Types

It is vital you pick approach research methodologies and methods for your thesis – your research after all is what your whole dissertation will rest on. When the researcher is willing to collect quantitative data it means that the variables are being measured and existing hypotheses are being verified and questioned. Often data is used … Continue reading “Comparison of Data Types”

Goal or Aim: The aim of qualitative research is more exploratory in nature. It provides a more detailed description of the research topic. On the other hand, quantitative research is more focussed on counting and classifying the features that comprise statistical models and figures which target to explain the observation.

Usage: In the beginning phase of the research qualitative research serves a better purpose and when one talks about the latter part of the research, quantitative research has more weightage. A clearer picture about what to expect from research is drawn from quantitative research vis a vis qualitative research.

Data Gathering: In the case of qualitative research, the main data gathering instrument is the researcher. The different strategies that the researcher employs depend largely upon the approach of the research. Some of the examples of the techniques in qualitative research are in depth interviews, structured unstructured interviews, narratives etc. When one talks of quantitative research tools, the instruments used are questionnaires and surveys to collect numerical data which is measurable.

Presentation of the Data: In the case where the data is of qualitative nature, in any of the forms such as words, images or objects, it appears in the form of graphical figures. On the other hand, if the research is quantitative in nature, the tabular representation of data is there which is in the form of numbers or statistics

Types of Study Design

Cross Sectional Studies: The cross sectional studies are simpler and target at discovering the phenomenon, concern, attitude by looking at snap shot view of the cross section of the population. This helps to obtain a more holistic picture which holds relevance with the time in which the study was conducted. To give an example, the … Continue reading “Types of Study Design”

Cross Sectional Studies: The cross sectional studies are simpler and target at discovering the phenomenon, concern, attitude by looking at snap shot view of the cross section of the population. This helps to obtain a more holistic picture which holds relevance with the time in which the study was conducted. To give an example, the use of a cross sectional design would be to have a look at the demographic characteristics of the population. The advantage of these kinds of studies are that they mostly undertake only one contact with the population under study and the cost involved in undertaking the study is relatively cheaper.

Some studies do pre-test and post-test work towards measuring the change in the situation or the phenomenon that is the focus of the study. These kinds of studies mostly measure the effectiveness of a programme. These kinds of studies are usually a variation from the conventional cross sectional studies as they take into consideration two sets of data collection which is cross sectional on both sides. The purpose is to determine if any change has occurred.

Retrospective Studies: Another kind of study is the retrospective study. The name itself is self-explanatory and as obvious it is used to study a specific phenomenon which has had an occurrence in the past. The probable approach adopted in these studies is the use of secondary data collection which reveals information and data on the previous studies and databases.

Prospective studies: In contrast to retrospective studies, the prospective study design targets to estimate the probability of the occurrence of an event in the coming future. They help to guess what would be the outcome of any event. Usually the general science experiments are of a prospective nature as the experimenter can only know the effect or impact of an experiment only after it has run its course.

Longitudinal Studies: This kind of studies stretch over a longer time period where data is to be collected repeatedly throughout the course of the study. The time span of the longitudinal studies could be from few months to several decades. These kinds of studies help to identify the correlation between the variables and mostly are not able to give out a causal relationship

Categories in Hypotheses

A hypothesis is a tool of quantitative studies. It is a tentative prediction regarding the relationship between the variables which are being studied. The key work that the hypotheses do is that it translates the research question into a prediction of the outcomes that can be expected from it. The entire research is done as … Continue reading “Categories in Hypotheses”

In order to be complete, it is important that a hypotheses includes these three main components:

The variables
The Population
The relationship

The key features of hypotheses are:

Stated clearly by using the appropriate terminology
Testable
It should be a clear about the relationship between the variables
It should be having definable, limited scope

There is more than one type of hypotheses. They are:

Simple Hypotheses: These hypotheses help to predict the relationship between a single independent variable (IV) on one side and a dependant variable(DV).
Complex Hypotheses: This kind of hypotheses helps to predict the relationship that is there between more than two or two independent variable and likewise two or more than two dependant variable.
Directional Hypotheses: These kinds of hypotheses are drawn from theory. These imply that the researcher is committed to a particular kind of outcome. These kind of hypotheses
Non-directional Hypotheses: These kinds of hypotheses are used when there is little or no theory or when the findings are contradictory to previous study. They may have impartial implication and do not stipulate the direction of the relationship.
Associative and causal hypotheses: These kind of hypotheses propose relationships between two variables. In this case when one variable changes the other one also changes.
Null Hypotheses: As the name is indicative, they are used when the researcher insists that there is no relationship between the variables or when the empirical data is inadequate to state any kind of hypotheses. Null hypotheses can be simple, complex, causal and associative.
Testable Hypotheses: It includes those variables that can be measured or have the capacity to be manipulated. Their task is to predict a relationship on the basis of data.

What are Projective Techniques?

Projective techniques are seeked as they help in going to the deep sub consciousness. The way they work is as follows: All those who are participants in the research that follows this technique, expects them to project their feelings and thoughts onto other things that may not be otherwise apparent. For instance, if Pepsi was … Continue reading “What are Projective Techniques?”

Projective techniques are seeked as they help in going to the deep sub consciousness. The way they work is as follows:

All those who are participants in the research that follows this technique, expects them to project their feelings and thoughts onto other things that may not be otherwise apparent. For instance, if Pepsi was a bird then which bird would it become? The most common kind of projective techniques used are:

Completion of sentences
Completion of cartoons
Stereotyping
Personification of brands

After completing the first step, the participants are expected to explain their answers. This stage in the technique is important and the participant plays a crucial role in fairly answering the “Why” in this question. It is important because the projective techniques work towards releasing the sub conscious and attempting to reveal the real explanation by the means of probing. For example here, if Pepsi is seen as a pigeon by the participant then the explanation in the second stage may be that it is dull, silly and a trend follower rather than a trend creator in the market.

There is a lot of fun element that is involved in the projective techniques. It is found to be breaking the mundane monotony of the research and researchers, academicians, clients, respondents often look forward to the use of this technique. At the same time, it is very important to take care while using these techniques, particularly with the measurement of abstract things. It may create confusion between the truth and the error. The sub conscious is something that is more deep and profound than the abstract and often researchers are found struggling with new words and people, more than often do not even think about the sub conscious.

The development of projective techniques happened in psychology. The reliability of these techniques has been a concern for the alternate disciplines of research. Another area where projective techniques pick the mainstream is market research. This is largely because the other techniques are not able to generate an insight into the factors that lead to brand preference and buying.

The ambiguity in the projective technique responses can be minimized using these techniques:

The probing technique is very useful
Use of multiple techniques to draw out the most common conclusion
Combining data of multiple people with same responses.