Question

1. Give a short definition for each of the 3 measurement levels that you can use in SPSS to characterize a variable. Also, give one example variable for each measurement level. 2. Describe the basic function of inferential statistics. What does it mean when we say a result is significant when we conduct a quantitative analysis? What do the different levels of significance mean? Use example(s) to support your answer.

Answer 1

Answer 1.

The three measurement levels used to characterize variables in SPSS are Scale, Ordinal and Nominal.

Scale : Scale or continuous variables are those with categories ordered in numeric terms as the variables have a metric value. In this case, computation of distance is accurate as the values are numeric. Example of scale variable is age in numbers, or salary in dollars.

Nominal : A variable is characterized as nominal when its value has categories with no ordering. Example, the Zip-code of states in a country or the colors available for a t-shirt.

Ordinal : Ordinal variables are characterized as those variables whose values convey some form of ordering. For example, the happiness level represented by High, Medium and Low.

Answer 2.

The basic function of inferential statistics is to ascertain greater information from the data acquired from a sample. Its a means to achieve information about a population based on the inferences drawn from a representative sample as well as the probability of accurately predicting the results using the sample data as the input.

Statistically significant results are those that are not mere occurrences of chance, but have a reasoning associated with it. This is conveyed through the use of null hypothesis. Whether a null hypothesis is rejected or not rejected depends on the significance of the variables. If my Null Hypothesis states that Variable X is not going to be a factor affecting the dependent variable Y, and through the p-value, I find that as per my level of significance, the variable X is well within that range, it can be inferred that the Variable X has a significance on impacting variable Y and Null Hypothesis is rejected.

With statistics playing a role in understanding the probability of the results predicted being accurate, it is essential to decide the level that is considered to be the threshold of significance. Therefore, depending on the different industries, level of significance thresholds ( values) differ based on the seriousness of the industry.

For example, healthcare industry has a high level of significance as the seriousness of an inaccurate result being predicted is high whereas in a consumer retail industry, quality assurance levels may not have a very high level of significance as one or two products being defective will not harm the business or the stakeholders.