Statistics vs statistic

NOTE: Statistics Class

Written By: Putri Yunisari

STATISTIC

Ø  Latin word ‘Status” English word “State”

Ø  Statistic is the conclusion of fact (data) in the form of number which arranging in a table, graph, diagram, etc.

Ø  Data is the information that is not verify yet. Researcher needs to analyze it before it informed or published. It is absolute but dynamic.

STATISTICS

Ø  It is knowledge that related to the way of data collecting, classifying, analyzing and discussion making.

Ø  A set of mathematical equation to analyze things

Ø  Science of gaining information from numerical and categorical data. 

Anderson and Bancroft:

“Statistics is the science, art and method which is the most effective to collect and interpret quantitative data. So that easy to predict the mistake, conclusion or estimation through an inductive reasoning based on mathematics opportunity”

Webster:

“Statistics are the classified fact representing the condition of the people in the state especially those fact which can be stated in number or in table of number of in tabular of classidied arrangements”

Statistic based on technique:

1.        Descriptive

2.        Inferential

Statistics based on parameters:

1.        Parametric

2.        Non-parametric

Statistics based on variable:

1.        Univariate statistics

2.        Multivariate statistics

Purpose of Statistics:

1.        To predict something

2.        To draw rational conclusion

3.        To find out a correlation

Characteristics of Statistics:

1.        Use number/ it numerically expressed.

2.        It is collected in a systemic manner/order

3.        It is the collection of fact

4.        It is used for a specific purpose.

5.        It is estimated according to a reasonable standard of accuracy

6.        It is objective

7.        It is universal due to it is used in many fields of science

Functions of Statistics:

1.        Collects and presents data in a systematic order

2.        Classifies data and simplifies the complex data

3.        Provides basic and technique of comparison

4.        Helps to study the relationship between phenomena

5.        Indicates the trend of behavior

6.        Helps to formulate the hypothesis and to test it

7.        Helps to draw rational conclusion

8.        Becomes a quality control for standardization

9.        Develops researchers’ knowledge/ someone’s knowledge

10.    Makes a comparison of something

11.    Simplifies the data and draws it in simple way.

Benefits of Statistics:

1.        In daily life, it provides information to be interpreted and processed

2.        It provides a technique for data registration on the exact

3.        To present data briefly

4.        In business field, it helps production planning program

5.        It provides method for design: planning and carrying out research studies, description and inference

Problems or Basic Matter in Statistics:

1.        Average

2.        Variability or dispersion

3.        Correlation

4.        Rogation

5.        Communication

6.        Description

Examples:

Ø  A teacher find out the average score of her/his students in order to know their quality

Ø  A teacher will say the intelligence of A class students is homogeny compared to B class students. So that there was a difference of intelligence

Ø  The teacher assumption of students who are good in mathematics will also good in physics, and chemistry.

Branches of Statistics:

1.    Descriptive Statistics

Ø It describe the condition of fact (data) through parameters such as mean, modus or distribution of frequency

Ø The data will display in either table or graph form

Ø It is used only to describe not to summarize thing

Ø What should be served are the measurement of central tendency and the measurement of spread deficiency standard

Ø Deductive and doesn’t involve the sample

Ø Formative (development)

Ø Does not analyze hypothesis

B.S Escentt and Skandal:

“A method for shorting and counting (data) to make it transparent”

Scopes:

1.      Distribution of frequency

2.      The measurement of central value

3.      Data presentation in the form of graphic

4.      Index number

5.      Time series/ time arrangement

2.    Inferential Statistics

Ø  A technique of statistics to make a conclusion according to a smaller sample becomes a big conclusion for a population

Ø  The information comes from descriptive statistics

Ø  It uses assumption because inferential statistics need sample of population. Ex: in a quick president election, and in manufacture.

Ø  Method that used are T-test, anova, anacova, structural quotation model, analysis of regression.

Ø  Inductive.

Scopes:

1.      Probability

2.      Data distribution

3.      Estimation parameter

4.      Hypothesis test

5.      Regression analysis

6.      Correlation analysis

Types of Statistics Inferential

1.      Parametric

Ø  Statistics that makes assumption about the parameters (defining properties) of the population distribution from which one’s data are down.

Ø  It is used for data of interval and ratio, it be based on normal distribution model and homogeny variable

2.      Non-parametric

Ø  Statistics that makes no such assumption or a null category

Ø  Used for qualitative data and heterogenic variable

Ø  Doesn’t follow any distribution model and the variable should not homogeny type.

Population

Ø  Group of object to be studied or in which an investigator is primarily interested during his/her research problem.

Ø  Group of data to be collected

Ø  It is not only the number of object or subject, but also the characteristics that studied

Finite population:

Ø  Population with identification number or the total of population is identified.

Ø  Ex: the number of students in a school, employee in a factory

Infinite population:

Ø  Unlimited population or the total population is unclear, or undetected, or unidentified.

Ø  Ex: citizen which change every times (unless by limitation it can be finite)

Ø  In fact, all human (human being and animal) is considered as infinite population

Sample:

Ø  Part of population

Ø  Smaller number of population

Sampling Technique/Processes

1.        Probability

Ø  Used to give similar opportunity to all elements of population

Ø  Homogeny only

Ø  Quantitative research

Ø  Types:

1.      Simple random sampling

2.      Proportionate satisfied random sampling

3.      Disproportionate satisfied random sampling

4.      Cluster sampling

2.        Non-Proability

Ø  Used to take sample without randoming

Ø  Reasonable

Ø  Types:

1.      Systematics sampling

2.      Quota sampling

3.      Accidental sampling

4.      Purposive sampling

5.      Saturated sampling

6.      Snowball sampling

Simple Random Sampling

Ø  A sampling technique that every item of population has equal chance of being selected as the sample. It is a fair sample selection method.

Ø  It can be done through ordinal method, lottery, or random numeral table

Ø  The unit of population should not too big

Advantage:

1.        A fair method to reduce any bias involved

2.        Easy to pick smaller sample from the larger population

3.        Researcher doesn’t need prior knowledge to use this method

4.        It is a very basic data collection method

Disadvantage:

1.        A sampling error can occur with a simple random if the sample does not end up accurately reflecting the population.

2.        It is supposed to represent. It can be bias if the researcher knows lots about the sample.

Steps:

1.      Prepare the list of population members initially

2.      Mark each member with specific number. Ex: by numbering

3.      Chose the samples. It can be done through lottery or random number tables or random number generator software.

Example:

An organization has 500 employees and we want to take sample of it. Firstly, we make a list of all employees’ name. Second, we assign a sequential number to each employee. Third, figure out what your sample size is going to be (in the case the sample size is 100). The last, use a random number generator to collect the sample.

Proportionated Random Sampling:

Ø  It is usually use for population with leveling

Ø  The aspect of the population is heterogenic. The population divided into subpopulation based on certain characteristics (it is called stratification)

Ø  It is a technique that count a sample based on comparison

Example:

1.        Indonesian citizen is heterogenic based on education, religion, income, and etc. therefore, the sample should take by considering the differences of population characteristic.

2.        The population is 130 people and the sample we wanted is 50 people with the characteristics are:

Elementary graduation 20 person            20/130 x 50 = 7.69 or 8

Junior high graduation is 40 people         40/130 x 50 = 15.38 or 15

Senior high graduation is 55 people         55/130 x 50 = 21.15 or 21

University graduation is 15 person          15/130 x 50 = 5.77 or 6

Disproportionate Random Sampling:

Ø  It is used for population with leveling but not too professional

Ø  The aspect of population is homogeny

Cluster Sampling:

Ø  It is used for cluster/ group of individual

Ø  It is used for a biggest data

Systematics Sampling:

Ø  Systematical method that uses interval in the sample collection

Ø  Method that involved the selection of elements from an ordered sampling frame

Steps:

1.        List the sample

2.        Decide number of sample to be collected

3.        Define the interval (k)

4.        Define the first number of interval randomly. It usually done through lottery

5.        Take sample from the initial first number that already choose

6.        Choose one number; automatically the next interval will follow.

Example:

There are 1200 Acehnese English Teachers and you want to choose 80 of them as the sample of your research. The first thing you should do is listing all the population. Then try to find the interval by dividing the population size with the sample size (in this case 1200/80=15). After finding your interval (15 as the interval), choose the starting random number from 1-15 number. Whatever number you chose will be your first element/sample(s), automatically another sample will easy to detect. In this case you may chose 11, so the next sample must be number 26, 41, 56, 71, 86 and soon. Illustration:

Main element (S) = 11, second element (S + k) = 11+15, third element (S + 2k) = 11+30, and soon.

Quota Sampling:

Ø  A technique for selecting sample that has certain characteristics to full fill the quota needed no matter what.

Ø  The sample chose accidentally.

Ø  This technique is used when sample is collected in a certain number

Advantages:

Ø  It is easy, faster, cheap and relevant with the research

Disadvantage:

Ø  It is not representative because of the conclusion made in general

Example:

We want to study a public health service in Depok with the sample decided is 500 people. When the sample is not full fill, yet, the research marked as unfinished work. If there are 5 people who conduct the research, they have to make sure the quota is filled.

Accidental sampling:

Ø  It’s also known as convenience sampling.

Ø  The sample is chosen because of easier to access.

Example:

Someone was chosen as the sample because as accidentally she/he was on the moment when researcher looking for the sample or both researcher and the person has known each other. in example, a researcher study about the cleanliness of the environment and then she/he interviews everyone that she/he meet in that place.

Purposive sampling:

Ø  Kinds of non-random sampling method which also known as judgmental sampling

Ø  There were characteristics in choosing the sample that was specific based on the purpose of a research

Ø  It is used when the sample can’t be took randomly

Advantages:

1.        It will be relevant with the research design

2.        Cheaper and easier

3.        The sample already fixed.

Disadvantages:

1.        There is no guaranty the sample will be representative

Steps:

1.      Make sure the criteria

2.      Define the population be based on the previous study

3.      Define the minimal number of the sample

4.      Chose the sample

Example:

Researcher wanted to know the achievement of students who join OSIS. So that the sample must be those with criteria is the member of OSIS.

Saturated sampling:

Ø  A sampling technique that take the whole population as the sample because of the number of population is too small

Snowball sampling:

Ø  A method for identifying, choosing and take samples in a network of clan of continuous relationship.

Ø  Sampling method in which sample are obtained through a rolling process one respondent to another, this method is usually used to explain social patterns or communication (socio-metrics) of a particular community.

Ø  A sampling technique starts by a small sample than becomes a large sample. It is done by patterning where the first sample is used as the tool to get the next sample. The first sample give the information to the researcher about the second possible sample and soon. The minimum first sample needed is 2-12 persons. The big size of the sample is >30 and medium is 10-30. The time needed is < 6 weeks or 6 weeks – 6 months

Advantage:

1.        It is effective to find out some issues that visible or clearly disclosed, to study a certain community, communication issues, and etc.