Lecture 1 - Introduction, Fundamental Terminology & Basic Concepts Flashcards
What is Statistics (when used as a single noun)?/Define statistics/What is the definition of statistics?
A mathematical technique by which data are organized, treated, and presented for interpretation and evaluation
The science that involves collecting, summarizing, analyzing, presenting and interpreting data. It provides the logical framework which enables the objective evaluation of research questions of interest
Art and science of using quantitative information (data) to gain understanding and to make informed decisions
Why is Statistics important?
Statistics is an integral part of scientific research as it allows scientists to estimate population parameters, evaluate hypotheses and make informed decisions based on the uncertain information provided by sample data
(can elaborate individually on estimation of population parameters, evaluation of hypotheses and the making of informed decisions)
2 Types of research methods
Quantitative and qualitative
5 characteristics of quantitative research
Test theories using numbers
Pre-determined methods
Close-ended questions
Numerical data and analyses
Statistical analysis and interpretation
5 characteristics of qualitative research
Test theories using language
Emerging methods
Open-ended questions
Text/image data
Identification of themes and patterns
What is measurement?
Process of collecting data i.e. measuring variables
Act of assessing, such as process of comparing a value to a standard, or counting frequency of occurrence of events
Quantitative or qualitative in nature
What is evaluation?
Philosophical process of determining the worth of the data
What is data?
Result of measurement
Set of measurements made on some part of the universe to address a particular information need or question
No mere collection of dis-joined, unrelated measurements obtained for no particular purpose
Plural
Each measurement is a datum or data point
What is a variable? Give an example
Refers to any characteristic of a person, place or an object that can assume more than one value/vary
e.g.
What is a constant? Give an example
A characteristic that can assume only one value and never changes
e.g. the distance of a running track
4 types/classification of variables [2 sub-classifications under categorical variables (see card 15)]
Quantitative and qualitative/categorical
Independent (predictor) and dependent (criterion)
What is a quantitative variable? Give an example
Naturally measured as a number for which meaningful arithmetic operations make sense
Variates differ in magnitude
e.g. Height, Weight, Age, Time, Distance, Temperature, Heart Rate, No.of students in a class, Numerical grades on an exam
What is a qualitative/categorical variable? Give an example
Any variable that is not quantitative
No numerical meaning when naturally measured
e.g. RPE scale, Sex, Race, Religion, Pain level on scale
2 ways that qualitative/categorical variable can be classified?
scale of measurement: ordinal and nominal scale
What is nominal scale? Give an example
Numerical values are just coding and in itself has no true numerical value
values function as labels rather than as numbers
Variates differ in category rather than in magnitude
e.g. Sex, Race, Religion
What is ordinal scale? Give an example
ordering in magnitude but interval between values is not interpretable and not equal
e.g. RPE scale, Pain level on scale
What is an independent/predictor variable? Give an example
Proposed cause of some effect
Predictor variable
Controlled/manipulated variable (in experiments)
e.g. training program
What is an independent/criterion variable? Give an example
Proposed/presumed effect
Dependent on the effects of one or more other variables
An outcome variable (that is observed as a result of the classification, control or manipulation)
Measured not manipulated (in experiments)
e.g. bench press 1RM
What is a population? Give an example
A well-defined group of individuals, places, objects, or observations of any size having a unique quality or characteristic
Eg: All social workers in Singapore
What is a sample? Give an example
A group of individuals, places, objects or observations selected from a particular population
Eg: Social workers in Singapore who volunteered for a research study
Why use a sample?
Limitations and constraints in terms of time, money i.e. not enough time, money, resources, equipment, manpower etc.
Just not possible to conduct experiment on the entire population
What is a parameter? Give an example
A characteristic of a population
Describes the population
Population parameter
e.g.
What is a statistic (when used a plural noun)? Give an example
A characteristic of a sample used to estimate the population parameter
Describes the sample
Sample statistic: estimate of a population parameter
e.g.
2 reasons for learning about the type of variables
- ) Decide how to interpret the data from that variable
e. g. Race (nominal): numerical values are just coding and in itself has no true numerical value
e. g. Race (nominal) or RPE (ordinal): equal differences between numbers do not imply equal differences in the amount of the attribute - ) Decide what statistical analysis is appropriate on the values that were assigned
Choice of descriptive or inferential statistical methods depends on the type of variables
What is validity/accuracy?
When designing an experiment the investigator must be certain that the study is technically sound i.e. the study must be valid and that one can draw meaningful and useful inferences from the results
Validity should be dealt with early on during the research design to prevent problems that may cast doubt on the implications derived from the results of the study
Related to systematic error
What is reliability/precision?
Ability of a measure to produce the consistent results under the same conditions
Consistency of a set of measurements
Extent to which measurement are free of random error i.e. random errors are related to/considered part of the reliability of a measurement
Related to:
Repeatability: the variation arising when all efforts are made to keep conditions constant by using the same laboratory, instrument and operator, and repeating during a short time period
Reproducibility: the variation arising using the same measurement process among different laboratories, instruments and operators over unspecified amount of time
What are the 3 things to look at when assessing validity and reliability in quantitative research?
measurement
sampling
study design
What is bias?
Any systematic (non-random) error in the design, conduct or analysis of a study
What are the 2 type of errors with regards to validity and reliability?
Systematic and random
Differentiate between systematic and random error
Systematic: caused by a specific factor in the design, conduct or analysis of a study; consistent in same direction; reproducible inaccuracies; considered part of the validity of a measurement
Random error: caused by unknown and/or unpredictable factors; inconsistent; considered part of the reliability of a measurement
Differentiate between sample statistics and population parameters
Population parameter is , while sample statistic is the estimate of a population parameter
Differentiate between validity and reliability
Validity is the technical soundness of the experiment and the ability of one to be able to draw meaningful and useful inferences from the results, while reliability is the ability of a measure to produce the consistent results under the same conditions
Validity of an experiment is related to systematic errors, while reliability is related to random errors
What are the 4 types of level of measurement scale involved? List from weakest to strongest/highest. Give some examples
Nominal: no ordering e.g. Sex, Race, Gender, Nationality, blood type
Ordinal: ordering exists, but not distance e.g. percentile ranks, grades at school, ranks in a race, letter grade in exam
Interval: distance exists, but not ratios e.g. temperature (in C, F, or R), water level of a river, dates (years), temperature in Celsius, IQ scale, numerical grade on exam
Ratio: ratios exist e.g. temperature in K, weight, driving speed, velocities, lengths, temperature in Kelvin, age