The data collected during the research project need to be processed and analyzed systematically so that trends and relationship patterns can be detected. Data analysis and interpretation enable the researcher to reduce, summarize, organize, evaluate, interpret and communicate numeric information in the descriptive form. The data analysis is planned while developing the research proposal, which allows the researcher to see how they will examine the data, Summarize the finding meaningfully, and Draw conclusions about the findings.
The term analysis refers to the statistical categorization, sorting, and summarization of data in order to gain answers to research questions. When the whole data set has been acquired, the researcher begins the data analysis process. Correct data organization may save much time and prevent mistakes.
Most researchers utilize a database or statistical analysis application to structure data to match their needs and properly organize their data. Each data analysis program has its criteria for data entry. Competent researcher inputs all the data in the same format and database since doing so may result in confusion and problems with the statistical analysis later. Interoperability concerns between various software might arise from time to time.
Data gathered from diverse sources using various instruments and procedures often consists of numerical numbers, ratings, narrations, replies to open-ended questions from a questionnaire or an interview schedule, quotes, field notes, and so on. Typically, two categories of data are recognized in educational research, and these are both qualitative and quantitative data.
Qualitative data consists of linguistic or symbolic resources. Qualitative data includes thorough descriptions of observed behaviors, individuals, settings, and events.
These data are gathered using a variety of methods and techniques, which include −
Observation
Interviews;
Questionnaires, opinionnaries, and inventories; and
Recorded data from newspapers and other documents kept by educational institutions, courts, clinics, government and non-government organizations, and so on.
These are narrations that show what individuals have stated about their experiences and interactions in natural environments in their own words. The qualitative data acquired through observation may consist of action patterns and verbal and nonverbal interaction among group members in their natural environments. A questionnaire's replies to open-ended questions are neither systematic nor standardized. However, such comments are extensive and complete, providing qualitative data from respondents' feelings, opinions, and experiences concerning the phenomenon under research.
Quantitative data is gathered using numerous instruments and tests based on the scale of measurement: nominal, ordinal, interval, or ratio. We commonly encounter nominal, ordinal, or interval measurement scales in educational research. In educational research, ratio scale measures are essentially non-existent. People's experiences are categorized and assigned number values based on standard replies. These statistics are limited in scope and cannot give the depth and detail that qualitative data provide. Quantitative data might be parametric or non-parametric.
Descriptive Statistics approaches are widely utilized in educational research to analyze quantitative data. The approaches are divided into two types: descriptive statistics and inferential statistics. Descriptive statistics characterize a sample or population's features (attributes), limiting generalization to the specific group (sample). There were no inferences drawn beyond this group, but inferential statistical procedures were utilized to derive generalizations beyond the sample with a known degree of accuracy.
We use averages to describe the overall properties of samples (groups). The economic status of groups is indicated by 'average income,' 'grade point averages appraise the student's performance in a class'', or the atmosphere of a city is reported by 'average temperature.' However, in statistical jargon, the term "average" does not have a definite meaning because there are several types of averages, only one of which may be adequate for expressing certain features of a sample group. Three of the numerous possible averages are commonly utilized in the study of educational data. Mean, Median and Mode are examples.
Central tendency measures describe the location on an ordered scale. Although these statistics are highly valuable in explaining the nature of a measure distribution, they will not assist the researcher in determining how the observation data will likely be distributed. Another type of statistic, the measure of variability, is employed for this, and it is also known as the dispersion or spread measure. There are various measures of variability, but we will focus on three of them in this unit: I range, (ii) average deviation, and (iii) standard deviation.
When we offer a test to a random sample of students taken from a population, the results are raw. A raw score has no relevance on its own and only has meaning when compared to some reference group or groups. In general, we make comparisons using the following measurements.
A bivariate data set is one in which we receive measurements for two variables for each participant. For example, we get bivariate data if we obtain results from Mathematics and Hindi tests for a set of schoolchildren. The key feature of bivariate data is that each group member can have one measure coupled with another. When studying bivariate data, we want to determine the degree of association between the variables.
To confine data gathering to specific specified responses or categories, quantitative measurement employs instruments that give a standardized framework. The variables that characterize phenomena are assigned numerical values and fall into specified categories. However, in other cases, it is impossible to break down a phenomenon into discrete components or variables that can be quantified. In such circumstances, the researcher considers the phenomena as a whole and thinks there is some quality to the phenomenon.
A qualitative technique is used when a researcher strives to maintain the whole of the phenomena while validating assertions about it. Using this method, the researcher attempts to capture what individuals say in their own words. The qualitative technique explains people's experiences in depth and allows the researcher to record and comprehend individuals based on their perspectives.
Qualitative data are "detailed descriptions" of settings, events, individuals, interactions, and observable behavior. These data are also accessible as "direct quotes" from people regarding their experiences, attitudes, beliefs, and views. The qualitative aspect of the verbal data obtained through surveys, observations, and interviews predominates. Data of a qualitative character can also include 'excerpts' or 'entire passages' from papers, correspondence records, and case histories. It should be remembered that thorough descriptions, verbatim quotations, and qualitative case documentation are all examples of raw data from empirical settings.
The examination of organized material accessible through thorough descriptions, direct quotations, or case documentation to find intrinsic facts is referred to as qualitative data analysis. These data are examined from as many perspectives as possible to discover new facts or reinterpret previously known or existing facts.
The following approaches are commonly employed in qualitative data analysis:
The information gathered via the use of various instruments on the selected samples is both quantitative and qualitative. Quantitative data are measured on nominal, ordinal, interval, or ratio scales. These data are divided into parametric and non-parametric groups. Non-parametric data are either enumerated or ranked, whereas parametric data are acquired using interval or ratio measurement scales. We employ a nominal scale for enumerated data and an ordinal scale for ranking data. The quantitative data is summarised in a 'frequency distribution,' It can be graphically represented using a histogram, a frequency polygon, and an ogive. The categories of descriptive statistical measures are central tendency, variability, relative positions, and relationship.
Physics
Biology
Chemistry
Mathematics
Economics
Psychology
Environmental Science
Social Studies
Fashion Studies
English