Normal curve

Normal curve is a distribution curve for some variable of its probability. A balanced normal curve has same mean, median or mode. The probability is measured as the area covered under the normal curve for the given value of statistical significance. Lower is the value of normal curve, higher is probability as the covered area is larger. For 5% statistical significance, the probability under the area is mean plus minus twice the standard deviation. When statistical significance is to be interpreted and applied then some cautions are to be taken like correlation is a statistic which helps in measuring the extent of association between the different variables but it has to be understood that it does not have any results about causality. The judgment of the researcher should also be taken care of as it could lead to bias. It is better to test for smaller statistical significance as results will be more applicable. Correlation is a statistic which helps in measuring the extent of association between the different variables. The correlation tells about the direction as well as the strength of the relationship existing between variables. The value of coefficient of correlation lies between -1 to 1. So, it tells about the simultaneous movement of different means. Factorial design is an experimental design where the researcher can measure the cause-effect relationship between two or more dependent and independent variables. The interrelationship between different relationship is shown using square for 2 variables and cube for three variables. In factorial designs, interaction is a condition which happens when the combined effect of independent variables is different than the sum of individual effect of these variables. It helps in determining that either the variables are working independently or their workings are dependent on each other. Inferential statistics is mostly the base of research in cases where population size is large. Here the inference about the population is drawn from analysis of the random sample selected from it. In research, statistical significance is the probability that tells about the fact that the difference in findings of studied group is due to chance or differences are real. The value of statistical significance is between 0 and 1. Here 1 means the difference is completely due to chance, and 0 means the difference is completely real. So, lower the value of statistical significance, better are the test results. In research, alpha is representation of level of significance. is the probability that tells about the fact that the difference in findings of studied group is due to chance or differences are real. The value of alpha is between 0 and 1. Here 1 means the difference is completely due to chance, and 0 means the difference is completely real. So, lower the value of statistical significance, better are the test results. It is important because it tells the about the probability of choosing a wrong finding i.e. higher the value of alpha, the researcher gets to know that higher is the probability of choosing wrong hypothesis correct. When judgments or decision are to be made from single measure using one sample, then two types of analyses could be done. The first type is that descriptive statistic i.e. percentage or mean of sample could be compared to that of population. The second way could be analyzing the internal characteristics of the sample. When percentage or mean of sample could be compared to that of population then, there are two analyses which could be performed. The first analysis is Z-test which could be performed when the sample size is 30 or more than 30. And the second method is t-test which could be performed when the sample size is less than 30. When proportion of sample could be compared to that of population then there is one way of analysis which could be performed. The analysis which could be done is test of proportions where Z-test is performed with different formula. Chi-square test helps in measuring the internal element of the answers to a single measure. The value of Chi-square is the measure which tells that the difference measured in observed distribution is actual one or because of chance. If the calculated value is bigger than critical table value then the difference is due to chance. It is used when the data to be analyzed against a single measure is frequency distribution because chi-square analysis is the most correct and appropriate analysis to perform on the data which is in form of frequency distribution. When decisions are to be made about a single element from two or more independent samples the F-test is used. This is an elaborate analysis which even looks into the dependency between the samples. Here the difference between the subgroups calculated is measured to be either statistically different or just by chance. Normally special software is used to perform these tests because of the large amount of data and their complex structure. Factorial design is an experimental design where the researcher can measure the cause-effect relationship between two or more dependent and independent variables. The interrelationship between different relationship is shown using square for 2 variables and cube for three variables. The strength of the design is that it helps researcher evaluate when more variables are in play. The weakness is that it is confusing when more variables are there and it becomes diagrammatically difficult to display when more than three variables are there.