SF Basic Terms in Statistics PDF

MEDICAL COLLEGES OF NORTHERN PHILIPPINES MATHEMATICS IN THE MODERN WORLD LEVEL I 1.1 INTRODUCTION Descriptive and inferential statistics are used to STATISTICS...

MEDICAL COLLEGES OF NORTHERN PHILIPPINES MATHEMATICS IN THE MODERN WORLD LEVEL I 1.1 INTRODUCTION Descriptive and inferential statistics are used to STATISTICS analyze data, obtain samples and make Provides procedure in data collection, inferences about the population. presentation, organization, and The tools used in descriptive statistics are interpretation to have meaningful idea measures of central tendency and dispersion. that is useful for decision making. The tools used in inferential statistics are 1.2 BRANCH OF STATISTICS hypothesis testing and regression analysis. DESCRIPTIVE STATISTICS Visit the link below for additional Totality of methods and treatment information about descriptive and employed in collection, description, and inferential statistics analysis of numerical data. https://datatab.net/tutorial/descriptive- Purpose of descriptive statistics is to tell inferential-statistics something about the particular group of observation. 1.3 PARAMETER AND STATISTIC Descriptive statistics helps to describe PARAMETER and organize known data using charts, Is a numerical index describing a bar graphs, etc. characteristic of a population. INFERENTIAL STATISTICS STATISTIC Logical process from sample analysis to Is a numerical index describing a generalization or conclusion about a characteristic of a sample. population. Examples of Parameters Inferential statistics aims at making 20% of U.S. senators voted for a specific inferences and generalizations about the measure. Since there are only 100 population data. senators, you can count what each of them voted. A pharmaceutical company gave 1,000 volunteers a new vaccine to determine the prevalence of body aches as a side effect. Of the volunteers, 65% developed body aches within 48 hours of receiving the vaccine. Researchers can confirm the accuracy of this percentage because they know how many people took part in the Examples of Descriptive and Inferential study and how many people reported Statistics body aches afterward. Therefore, this Some examples of descriptive and inferential percentage qualifies as a parameter. statistics are given below: Examples of Statistic Suppose the scores of 100 students 50% of people living in the U.S. agree belonging to a specific country are with the latest health care proposal. available. The performance of these Researchers can’t ask hundreds of students needs to be examined. This millions of people if they agree, so they data by itself will not yield any valuable take samples or part of the population results. However, by using descriptive and calculate the rest. statistics, the spread of the marks can be Eighty percent of college students in the obtained thus, giving a clear idea United States report eating pizza at least regarding the performance of each once a week. Because the number of student. college students in the U.S. constantly Now suppose the scores of the students changes and encompasses a significant of an entire country need to be portion of the population, researchers examined. Using a sample of, say 100 calculate this statistic by surveying a students, inferential statistics is used to sample of all college students in the U.S. make generalizations about the This makes their finding a statistic as population. opposed to a parameter. Important Notes on Descriptive and Inferential Statistics LESSON 1: BASIC TERMS IN STATISTICS Page 1 of 4  MEDICAL COLLEGES OF NORTHERN PHILIPPINES MATHEMATICS IN THE MODERN WORLD LEVEL I 1.4 SOURCES OF DATA 1.5 CONSTANT AND VARIABLE There are two main sources of data whether There are two major characteristics of objects, primary or secondary. people, or events whether constant or variable. PRIMARY DATA CONSTANT Are data come from an original source, A constant is a characteristic of objects, and are intended to answer specific people, or events that does not vary. research questions, can be taken from VARIABLE interview, mail- in questionnaire, survey, A variable is a characteristic of objects, or experimentation. people, or events that can take of Primary data is often reliable, authentic, different values. It can vary in quantity and objective in as much as it was (e.g., weight of people), or in quality collected with the purpose of addressing (e.g., hair color of people). Variables can a particular research problem. be classified in different ways. A common example of primary data is the data 1.6 TYPES OF DATA collected by organizations during market There are basically two types of data of random research, product research, and competitive variables yielding two types of data: qualitative analysis. This data is collected directly from its and quantitative. original source which in most cases are the QUALITATIVE VARIABLE existing and potential customers. describes qualities or characteristics. It is Most of the people who collect primary data are collected using questionnaires, government authorized agencies, investigators, interviews, or observation, and research-based private institutions, etc. frequently appears in narrative form. SECONDARY DATA QUANTITATIVE VARIABLE Are data that are taken from previously are used when a researcher is trying to recorded data, such as information in quantify a problem, or address the research conducted, industry financial "what" or "how many" aspects of a statements, business periodicals, and research question. It is data that can government reports. It can also be taken either be counted or compared on a electronically (e.g., via internet websites, numeric scale. compact disk, etc.). Quantitative variables can tell you “How Secondary data is the data that has been many," "how much," or "how often." collected in the past by someone else but made available for others to use. They are usually once primary data but become secondary when used by a third party. Secondary data are usually easily accessible to researchers and individuals because they are mostly shared publicly. For example, when conducting a research thesis, researchers need to consult past works done in this field and add findings to the literature review. Some other things like definitions and theorems are secondary data that are added to the thesis to be properly referenced and cited accordingly. Some common sources of secondary data include trade publications, government statistics, journals, etc. In most cases, these sources cannot be trusted as authentic. Visit the link below for additional Visit the link below for additional information about sources of data information about types of data https://keydifferences.com/difference- https://www.scribbr.com/methodology/ between-primary-and-secondary- qualitative-quantitative-research/ data.html LESSON 1: BASIC TERMS IN STATISTICS Page 2 of 4  MEDICAL COLLEGES OF NORTHERN PHILIPPINES MATHEMATICS IN THE MODERN WORLD LEVEL I 1.7 CLASSIFICATION OF VARIABLES Continuous Variable Variables can be classified into two according to Height of a person purpose whether experimental or mathematical. Age of a person EXPERIMENTAL CLASSIFICATION A researcher may classify variables according to Visit the link below for additional the function they serve in the experiment. information about mathematical INDEPENDENT VARIABLES classification Are variables controlled by the https://www.g2.com/articles/discrete- experimenter or researcher, an vs-continuous-data expected to have an effect on the behavior of the subjects. 1. 8 LEVELS OF MEASUREMENT DEPENDENT VARIABLES The four widely recognized levels of is the variable being tested and measurement- the nominal, ordinal, interval, measured in an experiment, and is and ratio. 'dependent' on the independent NOMINAL LEVEL variable. the data can only be categorized In an experiment, the researcher is looking for ORDINAL LEVEL the possible effect on the dependent variable the data can be categorized and that might be caused by changing the ranked independent variable. INTERVAL LEVEL the data can be categorized, ranked, and evenly spaced RATIO LEVEL Examples of Independent and Dependent the data can be categorized, ranked, Variables in Experiments evenly spaced, and has a natural zero For example, we might change the type of EXAMPLES information (e.g., organized or random) given NOMINAL ORDINAL INTERVAL RATIO to participants to see what effect this might City of Top 5 Test scores Height birth Olympic (e.g., IQ or have on the amount of information medallists exams) remembered. In this particular example the type of Sex Language Personality Age information is the independent variable ability inventories (because it changes) and the amount of Ethnicity Likert- Temperature Weight information remembered is the dependent type in Fahrenheit questions or Celsius variable (because this is being measured). Civil Temperature Visit the link below for additional Status in Kelvin information about experimental classification https://www.voxco.com/blog/independ ent-variables-and-dependent-variables/ MATHEMATICAL CLASSIFICATION Variables may also be classified in terms of the mathematical values they any take on within a given interval. CONTINUOUS VARIABLE a variable that can take an infinite and uncountable set of values Visit the link below for additional DISCRETE VARIABLE information about levels of measurement a variable that can assume only fixed https://prinsli.com/types-of-data/ number of distinct values Examples Discrete Variable Number of printing mistakes in a book Number of road accidents in New Delhi Number of siblings of an individual LESSON 1: BASIC TERMS IN STATISTICS Page 3 of 4  MEDICAL COLLEGES OF NORTHERN PHILIPPINES MATHEMATICS IN THE MODERN WORLD LEVEL I 1. 9 SAMPLING TECHNIQUES B. NON- RANDOM SAMPLING One of the most important steps in the research Sampling procedure where samples process is to select the sample of individuals selected in a deliberate manner with little who will participate as a part of the study. or no attention to randomization. Sampling refers to the process of selecting It is also called non- probability these individuals. sampling. A. RANDOM SAMPLING CONVINIENCE SAMPLING process whose members had an is a non-probability sampling technique equal chance of being selected from where samples are selected from the the population population only because they are it is also called probability conveniently available to the researcher sampling PURPOSIVE SAMPLING SIMPLE SAMPLING Researchers select the samples based It requires using randomly generated purely on the researcher’s knowledge numbers to choose a sample. More and credibility. In other words, specifically, it initially requires a sampling researchers choose only those people frame, a list or database of all members who they deem fit to participate in the of a population research study. SYSTEMATIC SAMPLING Judgmental or purposive sampling is not It is a very common technique in which a scientific method of sampling, and the you sample every kth element. For downside to this sampling technique is example, if you were conducting surveys that the preconceived notions of a at a mall, you might survey every 100th researcher can influence the results. person that walks in, for example. QUOTA SAMPLING STRATIFIED SAMPLING It is a non-probability sampling technique Sampling starts off by dividing a similar to stratified sampling. In this population into groups with similar method, the population is split into attributes. Then a random sample is segments (strata) and you have to fill a taken from each group. quota based on people who match the This method is used to ensure that characteristics of each stratum. different segments in a population are SNOWBALL SAMPLING equally represented. It is a non-probability sampling type that To give an example, imagine a survey is mimics a pyramid system in its selection conducted at a school to determine pattern. You choose early sample overall satisfaction. It might make sense participants, who then go on to recruit here to use stratified random sampling to further sample participants until the equally represent the opinions of sample size has been reached. students in each department. also known as referral, respondent- CLUSTER SAMPLING driven, or chain referral sampling Starts by dividing a population into Visit the link below for additional groups, or clusters. What makes this information about non- random sampling different that stratified sampling is that https://www.questionpro.com/blog/non each cluster must be representative of -probability-sampling/ the population. Then, you randomly selecting entire clusters to sample. For example, if an elementary school had five different grade eight classes, cluster random sampling might be used and only one class would be chosen as a sample, for example. LESSON 1: BASIC TERMS IN STATISTICS Page 4 of 4