Calculation and Measurement

PHASE calculation and measurement in quantitative approach always accompanies a variable (read: Variables and Hypothesis). When a variable has been defined, the next thing is to collect as much data as needed from these variables. The collection is done in various ways (read: Data Collection Techniques). In the process of data collection is exactly counting and measurements are made. How to do it refers to the operational definition set. Here are brief introduction about matters related to the calculation and measurement.

Operational definitions. Simply put, the operational definition can be described as instructions or how to calculate / measure a variable. For example, a researcher wants to know the level of adolescent social media usage in a city. Before doing research, he must first explain the operational definition of each variable used in the study, among others: the level of use and social media. He could define these two variables like this:
the level of use is the frequency and duration of adolescents to access social media in a day.
Social media is a personal account on facebook and twitter teenagers.

In this example, social media in the group of categorical variables. Therefore the researchers quite count, how many teenagers are using facebook and twitter. However, the variable rate of use, researchers need to measure the frequency (for example, how many times a day) and duration (how many minutes in a day).

In general, the level of measurement can be four: nominal, ordinal, interval, and ratio. The type of data obtained following the measurements made on the level of a variable. For example, a variable that is measured by using a ratio pengkuruan level, will generate data ratio, not the nominal data. Likewise, the nominal level of measurement, will never be able to produce the kind of ratio data.

Level Measurement. This typology was first introduced by S.S. Stevens in 1946. He split level to a size measurement: nominal, ordinal, interval, and ratio.

a). Nominal. This is the simplest measure because enough variables grouped into certain categories. I prefer to call it a counting technique, not a measurement. For example, the variable sex Janis. Respondents were grouped into two categories: male and female. Once grouped, the researchers then perform calculations. If no number is not hinted at a nominal rate of weight measurement. For example, male and female coded 1 is marked with the number 2. This does not mean female weight is greater than the male. The figures merely as a marker.

b). Ordinal. Slightly higher than the nominal, ordinal measurement already hinted at the difference between the value of the other variable. Such differences are not absolute, but relative. For example, a researcher wants to measure the level of exposure to television cartoon programs for children. Respondents who watch over 2 hours in a day, grouped in the high category. Who watched between 1 to 2 hours into the category of being. As for who watched less than one hour is considered to be in the low exposure tingak. High, medium, low and it is the ordinal level of measurement.

c). Interval. If more emphasis on the level ordinal, interval measurement add the clear distance. For example, the variable level of education. Respondents who just graduated at the primary school level, of different weight with respondents who completed his doctoral studies. Researchers may stretch kateroti educational level becomes: (0) no school, (1) Early Childhood and Kindergarten, (2) Primary School, (3) Junior High School, (4) Public High School, (5) Diploma , (6) Bachelor - Tier 1, (7) Master - Tier 2, and (8) Doctor - Strata 3. With this assessment category, the distance between the primary level and doctoral studies, is very clear.

d). Ratio. The highest level of measurement is the ratio. Here is not just a category, the order and the distance that can be found. Weighting the data can be seen as any measured variable has a value of the absolute reference point (ie zero). For example, a researcher wants to know the economic level of a newspaper reader. Researchers define the variable rate of the economy as a total income of the newspaper readers in units of dollars a month. From here there will be a reference value of Rp 0, - (Zero Rupiah). The difference between the data amount of income derived from one respondent to another respondent does not just show the order and distance, but also the weight.