Exploratory Data Analysis - Unit I

Practice flashcards covering definitions and concepts related to data types, measurement scales, data collection methods, and sampling from Unit 1 of Exploratory Data Analysis.

Population (or Universe)

The group of all items being studied and about which conclusions or inferences are drawn; it refers to the complete group of persons or objects for which the results of a statistical enquiry apply.

Sample

A collection of some, but not all, of the units of the population; it is a subset containing the characteristics of a population selected to represent it in a research study.

Measurement

A process of assigning numbers to objects or observations, usually considered in terms of their mathematical properties.

Nominal Scale

A measurement scale that allows classifying observations into different categories where items belong to distinct classes that cannot be quantified or rank ordered.

Ordinal Scale

A measurement scale that classifies observations into different classes that are ordered on a continuum, allowing items to be ranked although differences between adjacent ranks may not be equal.

Interval Scale

A scale that allows for rank ordering and quantifying the sizes of differences between items with a constant size interval, but lacks an absolute zero point.

Ratio Scale

A measurement scale that features an identifiable absolute zero point, identifying the direction and magnitude of differences and allowing for meaningful ratio comparisons.

Primary Data

Original data collected primarily for the purpose of a given enquiry, acting as the raw materials of the investigation.

Secondary Data

Data already collected by someone else for another purpose and available for the present study; these are considered finished products.

Direct Personal Observation

A collection method where the investigator collects data personally by asking or cross-examining the informant in a tactful and courteous manner.

Indirect Oral Investigation

A method used when informants are reluctant to supply information, where the investigator approaches witnesses or third parties who are in touch with the informant.

Mailed Questionnaire

A data collection method involving a list of questions pertaining to the enquiry posted to respondents who are expected to write answers in blank spaces.

Online Methods

Data collection using forms posted to respondents via the internet, utilizing tools like Google Form, Microsoft form, or Computer-assisted telephone interviewing (CATI).

Information through Agencies

A method where the investigator appoints agents or correspondents to collect and transfer data from a wide area on a regular basis.

Schedules

A method of data collection where proformas containing sets of questions are filled in by specially appointed and trained enumerators who visit the informants.

Pilot Survey

A replica or rehearsal of the main survey used to test the questionnaire and bring to light any weaknesses in the survey technique.

Probability Sampling (Random Sampling)

Methods in which every item in the universe has a known chance or probability of being chosen for the sample, remaining independent of the person making the study.

Non-probability Sampling (Non-random Sampling)

Methods that do not provide every item in the universe with a known chance of being included, making the selection process partially subjective.

Simple Random Sampling

A method involving random selection from an entire population so that each possible sample is equally likely to occur.

Stratified Sampling

A sampling method involving the division of a population into smaller sub-groups known as strata based on shared homogeneous attributes.

Systematic Sampling

A probability sampling method in which a sample with a fixed periodic interval is selected from a larger population.

Multistage Sampling (Cluster Sampling)

A procedure where random selection is made of primary, intermediate, and final units from a population in several stages.

Judgment Sampling

A non-probability sampling method where the choice of sample items depends exclusively on the judgement of the investigator.

Quota Sampling

A non-probability method where quotas are set according to specified characteristics (such as income or age) and interviewers are instructed to fulfill these quotas.

Convenience Sampling (Chunk)

A method of obtaining a sample by selecting population units based on convenience rather than probability or judgement.

Snowball Sampling (Chain Referral)

A non-probability technique where existing study subjects recruit future subjects from among their acquaintances.

Editing the Data

The process of scrutinizing obtained data for possible errors, inconsistencies, or irregularities before proceeding to tabulation.

Geographical or Spatial Data

Data that are collected and categorized according to geographical areas or locations.

Chronological or Time-Series Data

Data that are observed and recorded over a specific period of time, such as population changes between $1951$ and $1991$.