Introduction
The hypothesis of this study is to demonstrate that during this 21st century the attitude of the students toward learning has changed a lot. To test and prove this hypothesis there is going to be made a survey, composed of a questionnaire made with students. Their selection will be made by random sampling.
Sampling is a research methodology where one does not have to interview or collect data from the entire population. To have a good report the sample chosen should be representative of the general population. In this case, the area of drawing my samples will be in schools. I will distribute my prewritten questionnaires to students and tutors. The sample would include control groups as well as randomly selected groups. The target sample would include students in the age group of 10-17yrs from a total of 40 schools from four different areas. Out of the sample population, I will have 20% from junior high, 20% from senior high, and 60% from colleges and universities. In the said learning institutions, I will use the random selection method. This means that I will give the questionnaires randomly in the schools to ensure that I reduce any chances of biasness. The first part of the questionnaire will be composed of nominal variables like gender, ethnicity, age, etc.
Independent, dependent, and outcome variables of the survey
The two major variables that we will be trying to assess the relationship are the student’s attitude toward learning and 21st-century classroom teaching. The first would be the dependent variable and the second one the independent variable. The major outcome of this research would be to demonstrate a causal relationship between the two variables; namely, that change in 21st-century classroom teaching methodology has brought about changes in the attitude of students toward learning. Speaking in general terms, this would be predictive research as the main aim is “to generalize from an analysis by predicting certain phenomena based on hypothesized general relationships” (Hayashi, 2007). To be more specific, the two major variables can be divided into numerous minor variables, socio–economic indicators. These will be used to assess the relationship between the two major variables.
One socio-economic indicator/variable would be the use of computers in the nowadays classroom as a method of teaching. Computers are new devices unavailable to previous generations of students and have already proven to have an impact on human behavior. We aim to demonstrate that their usage in classrooms has an impact on students’ attitudes on learning as well.
A second indicator would be the use of interactive boards inside classrooms. The combination of interactive visual effects in teaching is something new and that forces the mind to respond differently than without it.
A third indicator would be the use of the blogosphere in research and learning. Prior generations had not the possibility to discuss and interact with each other regarding homework or an issue as fast and with as much information as today’s students can through the use of the internet. Blogging is one of the major forms of using the internet to enhance your learning and give it a new dimension of social interaction, and fun. Our guess is to prove that blogging as a way of teaching students has changed their attitude toward learning.
Multivariate analysis of the data and rationale
The next logical step after the execution of a survey is data consolidation. Although there are different methods of data consolidation, spreadsheets and Access databases are the more common ones. Data consolidation should be done accurately as it is going to have a strong bearing on the required response rate and thus, on the objective of the survey. It would take around 4-5 days (Hayashi, 2007). Multivariate analysis is a “form of statistics encompassing the simultaneous observation and analysis of more than one statistical variable” (Alreck and Settle, 2005). It is essential to use this methodology because making simultaneous analyses of many variables, allows us to have a clearer picture of the relationship between them. Multivariate analysis is in fact a composition of several analyses. Below are some examples of the pretest data and usage of descriptive analysis, estimated scale reliability, and recording to an index/summary scale.
Descriptive Statistics
Overleaf, we see examples of descriptive statistics ran on a nominal variable (gender, Table 1) and on interval data (“Computers are central to learning”, Tables 2 and 3).
Table 1: Example of Descriptive Statistic on a Nominal Variable.
Table 2.
Table 3: Descriptive Statistics for the Interval Scale “Computers are central to learning”.
Recoding to an Index/Summary Scale
Creating an index from selected items in the first segment of the questionnaire is a two-step procedure. First, one check that the scores are all oriented towards 5 = “Strongly agree” and that all other non-valid answers (missing, DK, NAP) are removed). Second, one employs the SPSS commands Transform/Compute variable to generate an additive index, called COMPSAVVY, with values from zero to 15 (Pallant, 2007). The resultant frequency table is shown below.
Table 4: The Three-Item Index Variable ‘Comp Savvy’: Computers are Fun and Key to Learning.
Estimated Scale Reliability
For the above scale, the coefficient of reliability or internal consistency that is Cronbach’s α is accessed via the SPSS ver. 16 Analyze-Scale-Reliability analysis, with the options for means and inter-item correlations selected.
Table 5: Result of Cronbach’s α Run for Three Items Comprising CompSavy2 Scale.
The result (Table 5 alongside) is distinctly lower than the 0.80 hurdles acceptable in most social science research (Alreck & Settle, 1985). This suggests mediocre internal consistency and, in turn, that items 1, 3, and 4 do not measure the same construct in the minds of the pretest respondents. This impression, reinforced by low values in the inter-item correlation matrix (table not shown) is not necessarily a terrible thing because it means the three items do measure different constructs or attitudes and therefore merit inclusion.
References
Alreck, P., and Settle, R. (2004). The Survey Handbook. NY: McGraw Hill.
Hayashi, T. (2007). The Possibility of Mixed-Mode Surveys in Sociological Studies. International Journal of Japanese Sociology 16(1), 51–63.