Embedded Files
quantitative data
quantitative data
"Quant" means "number"
Data must therefore consist of countable numbers
Data "bins" are countable, as specific as possible, and the space between one value and the next could feasible be recorded indefinitely
e.g. 4:00 AM, 6:00 AM, 8:00 AM, 10:00 AM, 12:00 AM, 2:00 PM
It is feasible to take a reading between 4:00 AM and 6:00 AM, say, at 5:00 AM, 4:30 AM, 4:15 AM, or even 4:01.52
When collecting primarily quantitative data, it is often referred to as Numerical Research
EXAMPLES: age, height, weight, income, attendance, speed, heart rate, handspan, leaf length, temperature, distance travelled, population, count of skills performed, etc.
In the fields of genetics and natural selection, quantitative analysis is always a representation of continuous variation.
This is because the variable is controlled by a wide range of genes and environmental influences, resulting in a RANGE of phenotypes between 2 extremes, such as in human height.
Quantitative data is ALWAYS presented visually as data points that can be connected with a line graph or line of best fit. This will frequently result in a trend that can be modeled with a formula or in a Normal distribution. e.g.:
qualitative data
qualitative data
"Qual" refers to "quality"
Data must therefore consist of descriptive/linguistic observations
Data "bins" are not countable but categorical, and the space between one value and the next could not be feasible calculated or exist in nature
Fox, Mouse, Pig, Chimp, Donkey
I cannot feasibly determine what animal exists halfway between a Fox and a Mouse or 3/4 of the way between a Pig and a Chimp
When collecting primarily qualitative data, it is often referred to as Categorical Research
EXAMPLES: gender, religion, marital status, native language, social class, qualifications, method of treatment, teaching approach, strategy, type of animal, tongue rolling, fingerprints, eye colour, blood type, etc.
In the fields of genetics and natural selection, qualitative analysis is always a representation of discontinuous (discrete) variation.
This is because the variable is controlled by a limited number of genes and not environmental influences, resulting in a discrete number of phenotypes with no intermediates, such as in blood type.
Qualitative data is ALWAYS presented visually as a histogram. Note that a histogram is not a "bar graph" or "bar chart" because although the bars are spaced equally, they are not touching because the intermediates do not exist as categories in nature. e.g.:
Most all studies are most effective, reliable, relevant, and useful for analysis when BOTH quantitative and qualitative data are procured in tandem.
ADDITIONAL DETAILS
GENERALIZED DEFINITION
Quantitative data refers to any information that can be quantified — that is, numeric. If it can be counted or measured, and assigned a numerical value, it is naturally quantitative.
Quantitative variables can answer "how many," "how much," or "how often,” even over long spans of time.
OBJECTIVE
Measure specific data with units
Benchmark and track data over time (trends)
Measure the incidence of motivations, attitudes, and perceptions
Cautiously predict future behavior
METHODS
Quantitative data collection methods focus on highly controlled approaches and numerical analysis.
Surveys involve clear and consistent questions of all participants.
Polls involve the assigning of a numeric value to how each person responded, or counting the number of responses.
Experiments have control groups to compare the experimental group to, so all quantitative data records can be evaluated and compared equitably across several trials/iterations.
ADVANTAGES
Universally accepted (numbers are the same everywhere)
Quick and easy to collect and from which to draw conclusions
Specific statistical tests make sense for specific data sets
Less potential for error and subjectivity (e.g. outliers excluded)
DISADVANTAGES
Cannot tell you the “full story” or “big picture” alone
Numbers can be inconclusive, and reproduction of the experiment may be difficult
ADDITIONAL DETAILS
GENERALIZED DEFINITION
Qualitative data is descriptive, expressed in terms of language rather than numerical values. It describes important details that are not countable. Words and labels are used to describe characteristics and traits you have observed and recorded.
Qualitative variables can answer “why,” and “how.”
OBJECTIVE
Understand the likely rationale underlying motivations, attitudes, and perceptions
Provide insight into a problem definition, providing hypotheses and language necessary for subsequent quantitative evaluation
METHODS
Qualitative research methods are more flexible and utilize open-ended questions.
Interviews can be used to explore attitudes and opinions regarding specific issues and areas to improve upon.
Focus groups allow for specialized qualitative data to be recorded and evaluated in real-time.
Experiments usually have both quantitative and qualitative observations so that we can discuss the implications of our results in a reliable and reproducible manner.
ADVANTAGES
Rich, in-depth insights exploring context and rationale
Provides a predictive element for continuous data collection
People change their minds, and having detail allows us to track how and when perspectives changed and give clues as to why
DISADVANTAGES
Language has more potential to be subjective or misinterpreted
Analyzed by grouping, so more potential for bias
Relies on the experience of the researcher, so it is not statistically representative of the dataset unless double-blind or A/B, for e.g.
2 QUANTITATIVE VARIABLE SCENARIO
Can be displayed on a single graph
Weight (quantitative) is graphed as the explanatory (independent) variable on the X-axis and height (also quantitative) is graphed as the dependent variable on the Y-axis
The height and weight of red circles are graphed as points
The average comparative height and weight of blue circles are graphed as a line of best fit
We can easily see the relationship between height and weight of both red and blue circles at the same time, which is convenient because the weight of both types of circles is likely causal for their height
2 QUALITATIVE VARIABLE SCENARIO
Can be displayed on a single graph
Red circles and blue circles are assessed for size and are assigned to the Big Circle or Small Circle categorical "bin"
In this scenario it is impossible for there to be a medium sized circle category
The count of circles that fit into each of these categories are presented. The count is on the Y-axis.
Because both groups fit into both categories discretely, the data can be stacked
Because there is no prerequisite maximum or 100% of respondents to display, these stacked bars still function as a standard histogram, but saves space.
Page updated
Google Sites
Report abuse