Scientific Data Analysis

A crucial step in conducting research is scientific data analysis. The researcher must find an appropriate scientific method that is reliable and valid to perform the data analysis. There are various scientific data analysis methods, which we will discuss below.

What are the levels of measurement?

The levels of measurement are also known as scales of measurement. Levels of measurement in statistics describe and classify types of variables and how to measure them. They are designed to help us understand how to interpret the data, what statistical test to use, and what information the data can give us. There are four levels of measurement in psychological research, and they are as follows:

Nominal

The nominal level of measurement in psychology consists essentially of measurements of 'named' or 'labelled data'. An example of a nominal level measurement question is 'What is your gender?'. So the answers 'male', 'female' and 'other' are forms of nominal data.

Ordinal

The ordinal level of measurement in psychology is categorical data, and the values have a fixed set or order. An example of a question with ordinal measurement is 'What socioeconomic class do you belong to?'. So the ordinal data could be 'working class, 'middle class', and 'upper- class'.

The order of the data is important because it shows that one has a lower/higher value than the other, but we cannot determine by how much they differ. This type of method is usually used for qualitative data.

Ratio

Ratio level of measurement in psychology is a type of data that is classified and ranked, there is a clear difference between one point and the next. Ratio data is measured with continuous data. This measure has an absolute value of 0, meaning that the numerical values cannot be less than 0. Examples of data that use a ratio measure are participant height, age, travel speed. Your height cannot be negative, your age cannot be less than 0, and you cannot be travelling at a minus speed.

Interval

Similar to ratio measurement, interval data is a type of data that can be classified and ranked, meaning there is a clear difference between one point and the next. This data is measured with continuous data. The difference between the two levels of measurement is that interval level data can be less than 0 (0 is not absolute). An example is temperature, which can be recorded at 0 and below.

How is scientific data collection and analysis performed?

Case studies use different methods of scientific data collection called triangulation. Because of this, there are several methods that researchers must use for scientific data analysis. Examples of the most common methods are described in the following section.

Observations

Observations are usually recorded and analysed by trained professionals. An example of an analysis procedure is tally counting. In this analysis procedure, two professionals watch the video and count together to observe a particular behaviour or pattern.

The crucial step is to have two trained individuals do the counting, ensuring that the results are reliable. If the results are similar, the study suggests high inter-rater reliability. Therefore, this would indicate that the scientific data analysis method is appropriate for the research study.

Interviews

Semi-unstructured interviews use open-ended and closed-ended questions to obtain quantitative and qualitative data. To analyse them, notes are taken from these transcripts, which are categorised by themes. This process is called thematic analysis. Data is usually reported by stating the themes and patterns identified and providing excerpts from the transcript as evidence. This form of analysis provides qualitative data.

It enables the researcher and the reader to understand the phenomena in depth. It can be classified as scientific data analysis as it relies on evidence to support the interpretation of the themes, concepts and patterns identified by the researcher. This increases the validity of the research findings.

Personal records

Personal records such as diaries and letters provide qualitative information. The technique of scientific data analysis is quite different from that used for quantitative data. This is because the simplest and most reliable method of quantitative data interpretation is statistical data. Statistical data can be applied to qualitative data that was quantified if the scientific data analysis method is content analysis.

Content analysis is a method of analysis used to identify words, themes, and concepts in qualitative data, such as diaries, and follows a similar protocol to thematic analysis. However, content analysis quantifies words, themes, and concepts to understand their meaning and relationship. The statistical tests used for quantitative data can then be used.