DATA
Data is raw material that needs to be processed to produce both qualitative and quantitative information that shows facts or phenomena.
Before the data is processed, it is necessary to first test the validity and reliability, both in terms of theoretical, content and empirical constructs.
DATA FUNCTION
To provide information about the characteristics or behavior of a phenomenon (population) that we are studying, both discrete and continuous, so that we can get an idea or can draw conclusions and make decisions based on data that has been collected and processed.
Data Type:
- Based on the source, the data is divided into:
- Primary Data : Data that is cultivated/obtained by the researcher
- Secondary Data : Data obtained from other people/agencies
- By Type:
- Numerical data (quantitative) → expressed in numerical terms (numbers).
- For example: Data per capita income, expenditure, price, distance, etc.
- Categorical Data (Qualitative) → classified based on certain categories/classes
- For example: Achieving and Unachieving Student Categories,
- Small, medium and large city categories,
- Based on the level of Scale/Measurement the following characteristics:
- Nominal : There is no order, the order does not show the level (rank). No starting point No difference
- For example : What is your favorite color : 1. Purple 2. Gray 3. Brown 4. White
- Ordinal : There is an order. the order indicates the level (rank). No starting point No difference
- For example: How was your learning achievement last semester? 1. Very Good 2. Good 3. Average 4. Poor 5. Very Poor Nominal and Ordinal scales are used in relation to categorical/qualitative data.
- Interval : There is an Order There is a Difference There is no starting point
- For example: Temperature or temperature: 0°C does not mean it has no temperature. ; Musical scale; IQ
- Ratio : There is an order There is a difference There is a starting point
- For example: Income (Rp. 135 245.23 per month): Income Rp. 0 means none (compare with 0°C in temperature)
Research Data Collection Methods:
- Observation (observation involves all senses (sight, hearing, smell, smell, taste))
- Interview
- The interview is divided into:
- Unstructured interview.
- This is a structured interview preparation step.
- The question posed is an attempt to explore the initial issue.
- Spontaneous nature of questions
- Structured interview.
- Questions have been prepared, because it has been designed what data/information is needed.
- Interview Type
- Direct interview (face to face).
- Indirect interview: for example by telephone or internet (on-line)).
- Questionnaire (List of Questions)
- Questionnaire is a list of written questions addressed to respondents. Respondents' answers to all questions in the questionnaire were then recorded/recorded.
- Questionnaires are an efficient data collection method if the researcher knows exactly what data/information is needed and how the variables that state the required information are measured.
- Physical Measurement
- Laboratory Experiment
DATA RECORD
To achieve completeness, accuracy and clarity of data, data recording must be equipped with:
- Data collector name
- Data collection date and time
- Data collection location
- Additional information on data/terms/respondents
DATA ANALYSIS
Data Analysis : is an effort to find and organize data systematically to increase the researcher's understanding of the case under study and present it as findings to others.
- The process of data analysis in qualitative research begins by examining all the data collected from various sources, namely from interviews, observations that have been written down in field notes, personal documents, official documents, pictures, photos and so on.
- Notes are divided into two, namely descriptive and reflective ones (Noeng Muhadjir.2000: 139).
- Descriptive notes present events rather than summaries.
- Reflective notes further convey the frame of mind, ideas and concerns of the researcher. More displaying the researcher's comments on the phenomena at hand.
- After reading, studying, and reviewing, the next step is to carry out data reduction by making abstractions. Abstraction is an attempt to make a summary of the core, processes and statements that need to be maintained so that they remain in it. The next step is to arrange in units and categorization and the last step is to interpret and or give meaning to the data.
DATA PRESENTATION
- Presentation of data includes collecting, organizing, summarizing and presenting data (data collection, organization, summarization, presentation)
- Presentation of descriptive data (descriptive = give an overview) in the form of:
- Table
- Diagram Pie
- Block Diagram
- Line Chart
- More pictures
VARIABLE OR VARIABLE (VARIABLE)
Types of Variables.
- Independent Variables or Fixed Variables are: a number of symptoms or factors or elements that determine or influence the presence or appearance of symptoms or research responses. These variables in the implementation of experiments or research are called treatments or factors.
- The dependent variable or the dependent variable is the response of a study or experiment, namely a number of symptoms or responses that arise because of the independent variable. For example, the difference in body weight of broiler chickens due to being given different types of feed. So: The independent variable is the type of feed and the bound hair is body weight.
- Control variable ( Controle Variable ) is: a number of symptoms or factors or elements that are intentionally controlled, or equated so as not to interfere with or affect the independent variable or the dependent variable. Controlling the effect means that this variable does not participate in determining whether or not there is a response to the research results. So it can be expected that the dependent variable that appears is purely the result of the independent variable or treatment. For example, in an experiment with broiler chickens with different types of feed, the strain of the chicken, its sex and the cage must be the same, so strain, sex and the chicken coop are called control variables.
- Side variables or Intervining Variables are : a number of symptoms that cannot be controlled, but their effect on the dependent variable or response can be calculated. Because this variable affects the independent variable, it will cause the dependent variable that appears impure due to the independent variable, so it is necessary to know how much influence this variable has. One way to calculate the effect is to do a deflection or grouping. For example: if we want to examine all the sexes of broiler chickens, we have to classify males and females, so gender is no longer a control variable but has been used as an intermediate variable.
- Error Variables or Extraneous Variables are: a number of symptoms that cannot be controlled and their effects cannot be calculated or eliminated their effects on independent variables and/or bound variables, these variables may originate from the sample conditions and may also be outside the sample. This variable will appear when the research takes place, this variable will affect the accuracy of the research. The existence of this variable can be seen in the magnitude of the square of the middle of the error, the larger the square of the middle of the error, the greater the effect of this variable.
Experimental design is made with respect to techniques in overcoming and controlling the diversity/variables that interfere with the actual effect of the treatment or factor that we examine or determine is called Environmental Design (Enviromental Design).
In order for the effect of the treatment to be clearly seen, the variety of responses caused by the state of the experimental material should not obscure or distort the appearance of the effect of the treatment. Therefore, the diversity of responses caused by environmental conditions and the state of the experimental material used needs to be taken into account or removed or monitored, so that the impact of the treatment can be reduced to a minimum.
Qualitative and Quantitative Variables
As an independent variable or dependent variable or a factor, it can be classified as a qualitative factor and a quantitative factor. Qualitative factors consist of nominal rating scale levels or levels that can actually be seen as certain values of special officials with continuous concentration, but do not provide a meaningful arrangement. While the quantitative factors scale ordinal, interval or rational size.
Quantitative factors with certain levels can be viewed as values of continuous density variables, called regression factors, not every ordinal scale factor is included in quantitative factors, sometimes it is treated as qualitative factors. The sex factor of livestock consisting of male, female and castration is a qualitative factor, while the dose of drug administration with levels of 0, 5, 10 and 15 ml is a quantitative factor.
The distance between the lowest level and the highest level of a gradient factor of the independent variable is called the range of interest. Although in this range only the effective level t is determined, the researcher is interested in studying the influence of these factors in the continuum to the extent of a predetermined attention span, in other words inference is intended to allow an intrapolation to be used. But not to do extra polarization. Because this is outside the specified attention span and does not guarantee the reliability of the experimental data.
The distance between two successive levels in an arrangement means that the gradient factor is called the distance between levels. In a treatment design, these distances between levels may or may not be uniform. Factors with uniformly spaced levels are also called equidistant factors, while non-uniform factors are called unequally spaced.
Doses for drug administration have equidistant levels, for example 0, 5, 10 and 15 ml, while those that are unequally spaced are 0, 6, 8, 9 and 10 ml.
Qualitative factors do not recognize the concept of distance between levels, while the distance between successive levels of factors with an unmeasured ordinal rating scale remains.
2.3. Response Variable Measurement Scale.
We know 4 scales that can be used to measure facts as a source of data are as follows:
1. Nominal Scale.
Nominal scale is the measurement of the lowest level, this occurs when numbers or other symbols are used to classify objects, people, animals or other objects. If numbers or other symbols are used to identify groups to which several objects can be included, then the numbers or symbols form a nominal scale (classification).
For example, let's say we classify livestock into large livestock, small livestock, poultry and various livestock. Similarly, the classification of livestock after being treated to die and recover.
In this case, the scale for measuring livestock type variables consists of 4 points, while healing consists of 2 points. The point of the scale is called a class or category.
2. Ordinal Scale (Rank).
Ordinal scale occurs when objects that are in a category of a scale are not only different from those objects, but also have a relationship with one another. more difficult, more mature and so on.
Measurements made on an ordinal scale are objects that are distinguished according to their similarities and according to their order. So a complete and orderly sequence or ranking can be made among the classes. For example, the incidence of a disease in pigs is often, often, sometimes and never.
3. Interval Scale.
Measurements on the interval scale are stronger than the ordinal scale, because measurements are achieved in addition to the equation and sequence, the distance (interval) between two different classes is also taken into account.
The interval scale is characterized by the same units of measurement and constants which give a real number for each pair of objects in the consecutive set. In this kind of measurement the ratio between arbitrary intervals is independent of the unit of measurement, and the interval scale has a zero point.
An example of an interval scale is temperature, for example temperature measurements with the Celsius and Fahrenheit scales, these two temperature measurements have a zero point and different units of measurement, but both provide the same information. as well as the percentage (0 – 100%). All ordinal scales that have a zero point and arbitrary units of measurement, with a range greater than or equal to 5 can be included in the interval scale.
4. Rational Scale
The rational scale of a scale in addition to having properties such as an interval scale, plus another property, namely a certain zero point. In a rational scale, the ratio of two points on an arbitrary scale is independent of the unit of measurement. An example of a rational scale is a scale for measuring weight, length, content (volume), including the number of people or the number of livestock and so on.