The post Difference Between Concepts and Variables in Research appeared first on Reading Craze.

]]>Concepts are highly subjective in nature and that makes it difficult to use them as they are in a research study. These subjective thoughts cannot be measured on a statistical scale. Kumar (2000) says that concepts are mental images and therefore their meanings vary markedly from individual to individual. Concepts are subjective impressions and their understanding will differ from person to person, which, if measured, would cause problems in comparing responses. Concepts should be converted into variables so that they can be measured, although on different scales same variable will have different precision.

If the researcher is using some concepts in his research he needs to find out some indicators that are reflective of these concepts. these indicators can be chosen subjectively by the researcher but they should have a logical link with the concept. The indicators can then be converted into variables.

- Take an example of the effectiveness of a medicine in curing a disease, the researcher can use the indicators: changes in the mortality rate, changes in morbidity, changes in recurrence of that disease, or prevention from that disease. These indicators can then be converted into variable to be able to be measured.
- We can take another example of another concept that is how rich someone can be? To measure this concept you need an indicator and you can measure someone’s richness from his wealth that he possesses. This wealth can be in the form of his income, money in his bank accounts, owned houses or other property and so on.
- A psychologist might want to test the effectiveness of his counselling to patients with anxiety. Effectiveness is a concept and you cannot measure it on any statistical scale. He can operationalize his concept of effectiveness of his counselling treatment into the following indicators: percentage reduction in patients’ anxiety, reduction in his day offs from the workplace, reduction in his visits to the psychologist office etc.

Without converting your concept into an indicator and then a variable you cannot measure it on any scale. The subjectivity of these indicators make them not suitable to be used directly in a research project. It should also be noted that the extent of variation can only be reduced by operationalizing these concepts, it cannot be eliminated completely.

Variables are measurable of course, with varying degree of accuracy. Measurability is the main difference between concepts and variables. A variable can be measured either using crude or refined method or either using subjective or objective methods. There are various scales and a variable can be measured on either one of those scales. The statistical variables can be measured on either nominal, ordinal, ratio or interval scale. This ability of the variables brings objectivity in the research findings.

A s variables are capable of measurement they can take different values and every variable can have different values. Generally speaking variables can be either independent variable or dependent variable. There can also be extraneous and intervening variables

From the viewpoint of causation an independent variable is a variable that affects the dependent variable and in itself it is free of any effects from the dependent variable. It is the cause for the change in any phenomenon, situation, disease etc. For example in testing the cause of juvenile delinquency in a community, availability of guns can be taken as the cause and hence the independent variable.

The dependent variable is the other main variable that is the effect of the independent variable. For example in a research on the impact of the availability of guns on the youth crime rate in a certain community, the youth crime rate is the dependent variable. In the above example the crime rate among youth is dependent on the availability of the guns.

In reality the situation is not always perfect with independent and dependent variable. Extraneous variables are all those variables that can impact the dependent variable other than the independent variable. In a laboratory setting it is comparatively easier to do the experiments in a perfect environment where the researcher controls all the extraneous variables. On the other hand, in a naturals setting it is difficult to control the extraneous variables.

In certain situation an intervening variable needs to be there to have the independent variable affect the dependent variable. this variable is not always present but in the certain situation its intervention plays an important role between the cause and effect relationship.

- Kumar, R.,
*Research Methodology: A Step-by-Step Guide for Beginners,*Sage Pub, London, 2000 - Kothari, C.R.,
*Research Methodology: Methods and Techniques,*New Delhi,Wiley Eastern Limited, 1985

The post Difference Between Concepts and Variables in Research appeared first on Reading Craze.

]]>The post Interval Scale in Research Methodology appeared first on Reading Craze.

]]>- This scale has constant equal distances between each successive values.
- This scale has wide scope due to its characteristic that it can categorize the data in equal intervals.
- Mean and standard deviation can be applied to data measured using this scale. The range can also be calculated to obtain the data dispersion.

- This scale has all the characteristics of a complete measuring scale except that it does not have an absolute zero.

Selection of the right statistical technique and data analysis depends heavily on the variables to be studied and the measurement scales used in the research. Therefore, the selection of right measuring scale is crucial to the success of the research. Data that has been measured using this scale can be treated by one of the several statistical techniques including mean, standard deviation, regression, correlation, range, analysis of variance etc. Studentized range and coefficient of variation cannot be calculated because ratios have no meanings in this scale.

- Campbell MJ, Machin D, Wiley J.
*Medical Statistics: A Commonsense Approach.*Vol. 2. London: Wiley; 1993. - Stevens SS. Bobbs-Merrill,
*On the Theory of Scales of Measurement.*College Division; 1946. - Marateb H. Reza, et al,
*Manipulating measurement scales in medical statistical analysis and data mining: A review of methodologies.*Journal of Research in Medical Sciences; 2014 Jan; 19(1): 47–56.

The post Interval Scale in Research Methodology appeared first on Reading Craze.

]]>The post Ordinal Scale in Research Methodology appeared first on Reading Craze.

]]>- It sub-categorizes values in the order of their magnitude. For example, the attitude of students towards a school program can be categorized in very effective, effective and not effective. The typical 5 or 7 point Likert scale is an example of ordinal scale. The use of Likert scale is very common in social sciences.
- This scale has greater scope for psychology and other social sciences research.

- This scale sorts the data in order but it does not quantify the relationship between different sub-categories.
- Because of lack of equal distances between scale points, arithmetic and statistical operations are not possible.

The researcher can apply median and mode to ordinal scale measurements but mean and standard deviation cannot be applied. Application of median is possible because data in ordinal scale is arranged in descending or ascending order. This arrangement of data allows finding out the median of the data where most values lie. On the other hand, in nominal scale, as data is not arranged in any order median cannot be applied. A median cannot be used in further mathematical calculations and therefore it cannot be used in statistical analysis. For nominal and ordinal scale mode can also be calculated and for nominal scale mode is the only value that can be calculated.

- Manikandan, S (2011). “Measures of central tendency: Median and mode”.
*Journal of Pharmacology and Pharmacotherapeutics*.**2**(3): 214, 215. doi:10.4103/0976-500X.83300. PMC 3157145. PMID 21897729. - Chrisman, Nicholas R. (1998). “Rethinking Levels of Measurement for Cartography”.
*Cartography and Geographic Information Science*.**25**(4): 231–242. doi:10.1559/152304098782383043. ISSN 1523-0406. Retrieved 14 August 2015. – via Taylor & Francis - Boone N. Harry (2012).
*Analyzing Likert Data. Journal of Extension*. April 2012 Volume 50 Number 2 Article Number 2TOT2. West Virginia University. - Gail M. Sullivan, MD, MPH and Anthony R. Artino, Jr, PhD.
J Grad Med Educ. 2013 Dec; 5(4): 541–542. doi: 10.4300/JGME-5-4-18.*Analyzing and Interpreting Data From Likert-Type Scales. Journal of Graduate Medical Education.*

The post Ordinal Scale in Research Methodology appeared first on Reading Craze.

]]>The post Types of Measurement Scales in Research and Statistics appeared first on Reading Craze.

]]>The nominal or classificatory scale is the most simple scale that is used in statistics and research. This scale only identifies the variables under study into unique values. This scale does not provide numerical values to the variables. The variables are only compared on the basis of some unique characteristic descriptively. Kumar R. (2000) defines nominal scale as the classifications of individuals, objects or responses based on common or shared property or characteristics. These people, objects or responses are divided into a number of subgroups in such a way that each member of the sub-groups has a common characteristic. This scale is most commonly used to categorize ‘gender’ or other simple variables, such variables do not have numerical values. Gender can be either male or female, but you cannot assign some numerical values to gender like 100 percent male etc.

The ordinal scale has the properties of the nominal scale, but it is a little more advanced than nominal scale. It categorizes values assigned to the variables on the basis of their magnitude. Some values related to the variables will be ‘smaller’ while other will be ‘larger’ or in other cases, some values will be ‘greater’ than the other values which will be ‘lesser’. The values are arranged either in ascending or descending order, so there is an ordered relationship between the values in the scale. This scale, however, does not categorize the values on the scale in fixed intervals.

The interval scale has the properties of the ordinal and nominal scale, which means that it assigns unique values to the each value under study and it also categorizes the values in ascending or descending order. The unique property of the interval scale that makes it different from the previous two scales is that it also categorize the values in equal intervals. In this scale, any of the two values on the scale has an equal interval. Temperature can be categorized using the interval scale, for example, values in centigrade can be spaced at equal interval using this scale. It also happens that in this scale the values have a starting point and an ending point. The space between 0-degree centigrade to 100-degree centigrade is always equally spaced.

This scale has all the properties of the previous scales plus one of its own properties that it has a fixed zero point, and below that point, no value exists. Every value can be measured from an absolute zero point or starting point, due to this property this scale is called as an absolute scale and for most empirical and mathematical operations this scale is used. Weight and height are the best examples of variables that can be measured on the ratio scale, they have an absolute zero point and no value exists below that. Temperature, which can be measured on the interval scale, cannot be measured on the ratio scale because the temperature can be below zero so there are no absolute zero points.

The post Types of Measurement Scales in Research and Statistics appeared first on Reading Craze.

]]>The post The Variables in Research Studies appeared first on Reading Craze.

]]>This is an important question that students should know before attempting to formulate a research problem. In general terms, variables are concepts that can be measured and operationalized. In the research, there are concepts that the researcher deals with these concepts are highly subjective, for example, some concepts about a community health program. There will be different concepts about the program, say it will be successful or it will a total failure. The term success or failure can be very subjective and judgemental. You can ask the respondents what do they think about this program and they will say it will be *effective, successful, failure, useful. *These terms and other such judgemental terms cannot be treated scientifically in the research. The reason being that their meaning can be very different from one person to another. The context in which each person looks at *success *or *failure* can be very different. One person might say that this particular community program benefited 50% of the community members and so it is successful but another person might say that at least 75 % people should benefit from this program to call it a success. This subjectivity can bring bias in the research. To avoid this subjectivity, the concepts should be converted into variables in research. Variables are quantifiable and hence they can be easily dealt in the research process.

Very simple definitions of variables say that a variable is something that can change. Variables are concepts that can be measured on any one of the measuring scales. Numeral values can be assigned to the variables. There are several other definitions of variables in research like variables are those aspects of the research that vary and this variation can be tested using different scales. These variables are characteristics or values and hence they can change, their use is more common in psychology but in other studies to they can be used.

Here we take two examples of variables in research and try to understand how variables influence the research process. Suppose you want to study the effect of permanent-press finishes on the durability characteristics of cotton fabric. It is an experimental study and in experiments you have an independent variable, a dependent variable and a control variable. The dependent variable, therefore, will be the cotton fabric and the independent variables are the permanent-press finishes that are applied to cotton fabric. There can be several other variables that can effect this study, these variables need to be controlled. For example, the other finishes that are applied to the cotton fabric can be the extraneous variables and they must be controlled to see the effect of independent variable on the dependent variable in the study.

Take another example, suppose you want to study the impact of the tsunami on the religious and social lives of the people living on the west coast of Sumatra-Indonesia. It is again a cause and effect study and in this study you will take the lives of the people as a dependent variable and tsunami after effects as the independent variables. Any other extraneous variable should be controlled in order to avoid bias in the study. One thing to make sure in such researches or studies is that the control variable needs to be controlled in the best possible manner since it can ruin the whole study. Another way is to add the other control variables in the study and they can be taken as secondary independent variables. When more variables are introduced in the study the study becomes more empirical and valid but it becomes complicated. It takes more time to measure several variables than to study the impact of one variable. When extraneous variables are difficult to control they should be added in the study.

The post The Variables in Research Studies appeared first on Reading Craze.

]]>