Research

Systematic investigative process employed to increase or revise current knowledge by

discovering new facts. It is divided into two general categories:

(1) Basic research is inquiry aimed at increasing scientific knowledge, and

(2) Applied research is effort aimed at using basic research for solving problems or developing

new processes, products, or techniques.

In the broadest sense of the word, the definition of research includes any gathering of data,

information and facts for the advancement of knowledge.

The Scientific Definition

The strict definition of scientific research is performing a methodical study in order to prove a

hypothesis or answer a specific question. Finding a definitive answer is the central goal of any

experimental process.

Research must be systematic and follow a series of steps and a rigid standard protocol. These

rules are broadly similar but may vary slightly between the different fields of science.

Scientific research must be organized and undergo planning, including performing literature

reviews of past research and evaluating what questions need to be answered.

Any type of ‘real’ research, whether scientific, economic or historical, requires some kind of

interpretation and an opinion from the researcher. This opinion is the underlying principle, or

question, that establishes the nature and type of experiment.

The scientific definition of research generally states that a variable must be manipulated,

although case studies and purely observational science do not always comply with this norm.

Empirical Research

Empirical Research can be defined as "research based on experimentation or observation

(evidence)". Such research is conducted to test a hypothesis.

The word empirical means information gained by experience, observation, or experiment. The

central theme in scientific method is that all evidence must be empirical which means it is based

on evidence. In scientific method the word "empirical" refers to the use of working hypothesis

that can be tested using observation and experiment.

Empirical data is produced by experiment and observation.

Objectives of the Scientific Research Process

Capture contextual data and complexity

Identify and learn from the collective experience of others from the field

Identification, exploration, confirmation and advancing the theoretical concepts.

Further improve educational design

Objectives of the Empirical Research

Go beyond simply reporting observations

Promote environment for improved understanding

Combine extensive research with detailed case study

Prove relevancy of theory by working in a real world environment (context)

Reasons for Using Empirical Research Methods

Traditional or superstitional knowledge has been trusted for too long

Empirical Research methods help integrating research and practice

Educational process or Instructional science needs to progress

Advantages of Empirical Methods

Understand and respond more appropriately to dynamics of situations

Provide respect to contextual differences

Help to build upon what is already known

Provide opportunity to meet standards of professional research

In real case scenario, the collection of evidence to prove or counter any theory involves planned

research designs in order to collect empirical data. Several types of designs have been

suggested and used by researchers. Also accurate analysis of data using standard statistical

methods remains critical in order to determine legitimacy of empirical research.

Various statistical formulas such uncertainty coefficient, regression, t-test, chi-square and

different types of ANOVA (analysis of variance) have been extensively used to form logical and

valid conclusion.

However, it is important to remember that any of these statistical formulas don't produce proof

and can only support a hypothesis, reject it, or do neither.

Empirical Cycle

Empirical cycle consists of following stages:

1. Observation

Observation involves collecting and organizing empirical facts to form hypothesis

2. Induction

Induction is the process of forming hypothesis

3. Deduction

Deduct consequences with newly gained empirical data

4. Testing

Test the hypothesis with new empirical data

5. Evaluation

Perform evaluation of outcome of testing

The Scientific Definition

The strict definition of scientific research is performing a methodical study in order to

prove a hypothesis or answer a specific question. Finding a definitive answer is the

central goal of any experimental process.

Research must be systematic and follow a series of steps and a rigid standard protocol.

These rules are broadly similar but may vary slightly between the different fields of

science.

Scientific research must be organized and undergo planning, including performing literature

reviews of past research and evaluating what questions need to be answered.

Any type of ‘real’ research, whether scientific, economic or historical, requires some kind of

interpretation and an opinion from the researcher. This opinion is the underlying principle, or

question, that establishes the nature and type of experiment.

The scientific definition of research generally states that a variable must be manipulated,

although case studies and purely observational science do not always comply with this norm.

What is the Scientific Method?

Martyn Shuttleworth 194.4K reads 1 Comment

The scientific method, as defined by various scientists and philosophers, has a fairly rigorous

structure that should be followed.

In reality, apart from a few strictly defined physical sciences, most scientific disciplines have to

bend and adapt these rules, especially sciences involving the unpredictability of natural

organisms and humans.

In many ways, it is not always important to know the exact scientific method, to the letter, but

any scientist should have a good understanding of the underlying principles.

If you are going to bend and adapt the rules, you need to understand the rules in the first place.

Empirical

Science is based purely around observation and measurement, and the vast majority of

research involves some type of practical experimentation.

This can be anything, from measuring the Doppler Shift of a distant galaxy to handing out

questionnaires in a shopping center. This may sound obvious, but this distinction stems back to

the time of the Ancient Greek Philosophers.

Cutting a long story short, Plato believed that all knowledge could be reasoned; Aristotle that

knowledge relied upon empirical observation and measurement.

This does bring up one interesting anomaly. Strictly speaking, the great physicists, such as

Einstein and Stephen Hawking, are not scientists. They generate sweeping and elegant theories

and mathematical models to describe the universe and the very nature of time, but measure

nothing.

In reality, they are mathematicians, occupying their own particular niche, and they should

properly be referred to as theoreticians.

Still, they are still commonly referred to as scientists and do touch upon the scientific method in

that any theory they have can be destroyed by a single scrap of empirical evidence.

The Scientific Method Relies Upon Data

The scientific method uses some type of measurement to analyze results, feeding these

findings back into theories of what we know about the world. There are two major ways of

obtaining data, through measurement and observation. These are generally referred to as

quantitative and qualitative measurements.

Quantitative measurements are generally associated with what are known as ‘hard' sciences,

such as physics, chemistry and astronomy. They can be gained through experimentation or

through observation.

For Example:

At the end of the experiment, 50% of the bacteria in the sample treated with penicillin

were left alive.

The experiment showed that the moon is 384403 km away from the earth.

The pH of the solution was 7.1

As a rule of thumb, a quantitative unit has a unit of measurement after it, some scientifically

recognized (SI) or SI derived unit. Percentages and numbers fall into this category.

Qualitative measurements are based upon observation and they generally require some type of

numerical manipulation or scaling.

As an example, a social scientist interviewing drug addicts in a series of case studies, and

documenting what they see, is not really performing science, although the research is still

useful.

However, if he performs some sort of manipulation, such as devising a scale to assess the

intensity of the response to specific questions, then he generates qualitative results.

On average, the subjects showed an anxiety level of four.

91% of respondents stated that they preferred Hershey bars.

Generally, qualitative measurements are arbitrary, a scale designed to measure abstract

responses and constructs. Measuring anxiety, preference, pain and aggression are some

examples of concepts measured qualitatively. For a small group of long-established tests, the

results are often regarded as quantitative, such as IQ (Intelligence Quotient) and EQ (Emotional

Quotient).

Both types of data are extremely important for understanding the world around us and the

majority of scientists use both types of data.

A medical researcher might design experiments to test the effectiveness of a drug, using a

placebo to contrast.

However, she might perform in depth case studies on a few of the subjects, a pilot study, to

ensure that her experiment has no problems.

The Scientific Method is Intellectual and Visionary

Science requires vision, and the ability to observe the implications of results. Collecting data is

part of the process, and it also needs to be analyzed and interpreted.

However, the visionary part of science lies in relating the findings back into the real world. Even

pure sciences, which are studied for their own sake rather than any practical application, are

visionary and have wider goals.

The process of relating findings to the real world is known as induction, or inductive reasoning,

and is a way of relating the findings to the universe around us.

For example, Wegener was the first scientist to propose the idea of continental drift. He noticed

that the same fossils were found on both sides of the Atlantic, in old rocks, and that the

continental shelves of Africa and South America seemed to fit together.

He induced that they were once joined together, rather than joined by land bridges, and faced

ridicule for his challenge to the established paradigm. Over time, the accumulated evidence

showed that he was, in fact, correct and he was shown to be a true visionary.

Science Uses Experiments to Test Predictions

This process of induction and generalization allows scientists to make predictions about how

they think that something should behave, and design an experiment to test it.

This experiment does not always mean setting up rows of test tubes in the lab or designing

surveys. It can also mean taking measurements and observing the natural world.

Wegener's ideas, whilst denigrated by many scientists, aroused the interest of a few. They

began to go out and look for other evidence that the continents moved around the Earth.

From Wegener's initial idea of continents floating through the ocean floor, scientists now

understand, through a process of prediction and measurement, the process of plate tectonics.

The exact processes driving the creation of new crust and the subduction of others are still not

fully understood but, almost 100 years after Wegener's idea, scientists still build upon his initial

work.

Null Hypothesis

Martyn Shuttleworth 592.7K reads 9 Comments

The null hypothesis, H0, is an essential part of any research design, and is always tested, even

indirectly.

The simplistic definition of the null is as the opposite of the alternative hypothesis, H1, although

the principle is a little more complex than that.

The null hypothesis (H0) is a hypothesis which the researcher tries to disprove, reject or nullify.

The 'null' often refers to the common view of something, while the alternative hypothesis is what

the researcher really thinks is the cause of a phenomenon.

The simplistic definition of the null is as the opposite of the alternative hypothesis, H1, although

the principle is a little more complex than that.

The null hypothesis (H0) is a hypothesis which the researcher tries to disprove, reject or nullify.

The 'null' often refers to the common view of something, while the alternative hypothesis is what

the researcher really thinks is the cause of a phenomenon.

An experiment conclusion always refers to the null, rejecting or accepting H0 rather than H1.

Despite this, many researchers neglect the null hypothesis when testing hypotheses, which is

poor practice and can have adverse effects.

Examples of the Null Hypothesis

A researcher may postulate a hypothesis:

H1: Tomato plants exhibit a higher rate of growth when planted in compost rather than in soil.

And a null hypothesis:

H0: Tomato plants do not exhibit a higher rate of growth when planted in compost rather than

soil.

It is important to carefully select the wording of the null, and ensure that it is as specific as

possible. For example, the researcher might postulate a null hypothesis:

H0: Tomato plants show no difference in growth rates when planted in compost rather than soil.

There is a major flaw with this H0. If the plants actually grow more slowly in compost than in soil,

an impasse is reached. H1 is not supported, but neither is H0, because there is a difference in

growth rates.

If the null is rejected, with no alternative, the experiment may be invalid. This is the reason why

science uses a battery of deductive and inductive processes to ensure that there are no flaws in

the hypotheses.

Many scientists neglect the null, assuming that it is merely the opposite of the alternative, but it

is good practice to spend a little time creating a sound hypothesis. It is not possible to change

any hypothesis retrospectively, including H0.

Significance Tests

If significance tests generate 95% or 99% likelihood that the results do not fit the null

hypothesis, then it is rejected, in favor of the alternative.

Otherwise, the null is accepted. These are the only correct assumptions, and it is incorrect to

reject, or accept, H1.

Accepting the null hypothesis does not mean that it is true. It is still a hypothesis, and must

conform to the principle of falsifiability, in the same way that rejecting the null does not prove the

alternative.

Perceived Problems With the Null

The major problem with the H0 is that many researchers, and reviewers, see accepting the null

as a failure of the experiment. This is very poor science, as accepting or rejecting any

hypothesis is a positive result.

Even if the null is not refuted, the world of science has learned something new. Strictly

speaking, the term ‘failure’, should only apply to errors in the experimental design, or incorrect

initial assumptions.

Development of the Null

The Flat Earth model was common in ancient times, such as in the civilizations of the Bronze

Age or Iron Age. This may be thought of as the null hypothesis, H0, at the time.

H0: World is Flat

Many of the Ancient Greek philosophers assumed that the sun, moon and other objects in the

universe circled around the Earth. Hellenistic astronomy established the spherical shape of the

earth around 300 BC.

H0: The Geocentric Model: Earth is the centre of the Universe and it is Spherical

Copernicus had an alternative hypothesis, H1 that the world actually circled around the sun, thus

being the center of the universe. Eventually, people got convinced and accepted it as the null,

H0.

H0: The Heliocentric Model: Sun is the centre of the universe

Later someone proposed an alternative hypothesis that the sun itself also circled around the

something within the galaxy, thus creating a new H0. This is how research works - the H0 gets

closer to the reality each time, even if it isn't correct, it is better than the last H0.

Systematic and Methodical

Scientists are very conservative in how they approach results and they are naturally very

skeptical.

It takes more than one experiment to change the way that they think, however loud the

headlines, and any results must be retested and repeated until a solid body of evidence is built

up. This process ensures that researchers do not make mistakes or purposefully manipulate

evidence.

In Wegener's case, his ideas were not accepted until after his death, when the amount of

evidence supporting continental drift became irrefutable.

This process of changing the current theories, called a paradigm shift, is an integral part of the

scientific method. Most groundbreaking research, such as Einstein's Relativity or Mendel's

Genetics, causes a titanic shift in the prevailing scientific thought.

Summary

The scientific method has evolved, over many centuries, to ensure that scientists make

meaningful discoveries, founded upon logic and reason rather than emotion.

The exact process varies between scientific disciplines, but they all follow the above principle of

observe - predict - test - generalize.

Purpose of Research

The purpose of research can be a complicated issue and varies across different scientific fields

and disciplines. At the most basic level, science can be split, loosely, into two types, 'pure

research' and 'applied research'.

Both of these types follow the same structures and protocols for propagating and testing

hypotheses and predictions, but vary slightly in their ultimate purpose.

An excellent example for illustrating the difference is by using pure and applied mathematics.

Pure maths is concerned with understanding underlying abstract principles and describing them

with elegant theories. Applied maths, by contrast, uses these equations to explain real life

phenomena, such as mechanics, ecology and gravity.

Pure Scientific Research

Some science, often referred to as 'pure science', is about explaining the world around us and

trying to understand how the universe operates. It is about finding out what is already there

without any greater purpose of research than the explanation itself. It is a direct descendent of

philosophy, where philosophers and scientists try to understand the underlying principles of

existence.

Whilst offering no direct benefits, pure research often has indirect benefits, which can contribute

greatly to the advancement of humanity.

For example, pure research into the structure of the atom has led to x-rays, nuclear power and

silicon chips.

Applied Scientific Research

Applied scientists might look for answers to specific questions that help humanity, for example

medical research or environmental studies. Such research generally takes a specific question

and tries to find a definitive and comprehensive answer.

The purpose of research is about testing theories, often generated by pure science, and

applying them to real situations, addressing more than just abstract principles.

Applied scientific research can be about finding out the answer to a specific problem, such as 'Is

global warming avoidable?' or 'Does a new type of medicine really help the patients?'

Generating Testable Data

However, they all involve generating a theory to explain why something is happening and using

the full battery of scientific tools and methods to test it rigorously.

This process opens up new areas for further study and a continued refinement of the

hypotheses.

Observation is not accurate enough, with statistically testable and analyzable data the only

results accepted across all scientific disciplines. The exact nature of the experimental process

may vary, but they all adhere to the same basic principles.

Scientists can be opinionated, like anybody else, and often will adhere to their own theories,

even if the evidence shows otherwise. Research is a tool by which they can test their own, and

each others' theories, by using this antagonism to find an answer and advance knowledge.

The purpose of research is really an ongoing process of correcting and refining hypotheses,

which should lead to the acceptance of certain scientific truths.

Whilst no scientific proof can be accepted as ultimate fact, rigorous testing ensures that proofs

can become presumptions. Certain basic presumptions are made before embarking on any

research project, and build upon this gradual accumulation of knowledge.

Research Hypothesis

A research hypothesis is the statement created by researchers when they speculate upon the

outcome of a research or experiment.

Every true experimental design must have this statement at the core of its structure, as the

ultimate aim of any experiment.

The hypothesis is generated via a number of means, but is usually the result of a process of

inductive reasoning where observations lead to the formation of a theory. Scientists then use a

large battery of deductive methods to arrive at a hypothesis that is testable, falsifiable and

realistic.

The precursor to a hypothesis is a problem, usually framed as a question.

The precursor to a hypothesis is a research problem, usually framed as a question. It might ask

what, or why, something is happening.

For example, to use a topical subject, we might wonder why the stocks of cod in the North

Atlantic are declining. The problem question might be ‘Why are the numbers of Cod in the North

Atlantic declining?’

This is too broad as a statement and is not testable by any reasonable scientific means. It is

merely a tentative question arising from literature reviews and intuition. Many people would think

that instinct and intuition are unscientific, but many of the greatest scientific leaps were a result

of ‘hunches’.

The research hypothesis is a paring down of the problem into something testable and falsifiable.

In the aforementioned example, a researcher might speculate that the decline in the fish stocks

is due to prolonged over fishing. Scientists must generate a realistic and testable hypothesis

around which they can build the experiment.

This might be a question, a statement or an ‘If/Or’ statement. Some examples could be:

Is over-fishing causing a decline in the stocks of Cod in the North Atlantic?

Over-fishing affects the stocks of cod.

If over-fishing is causing a decline in the numbers of Cod, reducing the amount of

trawlers will increase cod stocks.

These are all acceptable statements and they all give the researcher a focus for constructing a

research experiment. Science tends to formalize things and use the ‘If’ statement, measuring

the effect that manipulating one variable has upon another, but the other forms are perfectly

acceptable. An ideal research hypothesis should contain a prediction, which is why the more

formal ones are favored.

A hypothesis must be testable, but must also be falsifiable for its acceptance as true science.

A scientist who becomes fixated on proving a research hypothesis loses their impartiality and

credibility. Statistical tests often uncover trends, but rarely give a clear-cut answer, with other

factors often affecting the outcome and influencing the results.

Whilst gut instinct and logic tells us that fish stocks are affected by over fishing, it is not

necessarily true and the researcher must consider that outcome. Perhaps environmental factors

or pollution are causal effects influencing fish stocks.

A hypothesis must be testable, taking into account current knowledge and techniques, and be

realistic. If the researcher does not have a multi-million dollar budget then there is no point in

generating complicated hypotheses. A hypothesis must be verifiable by statistical and analytical

means, to allow a verification or falsification.

In fact, a hypothesis is never proved, and it is better practice to use the terms ‘supported’ or

‘verified’. This means that the research showed that the evidence supported the hypothesis and

further research is built upon that.

A research hypothesis, which stands the test of time, eventually becomes a theory, such as

Einstein’s General Relativity. Even then, as with Newton’s Laws, they can still be falsified or

adapted.

True Experimental Design

True experimental design is regarded as the most accurate form of experimental research, in

that it tries to prove or disprove a hypothesis mathematically, with statistical analysis.

For some of the physical sciences, such as physics, chemistry and geology, they are standard

and commonly used. For social sciences, psychology and biology, they can be a little more

difficult to set up.

For an experiment to be classed as a true experimental design, it must fit all of the following

criteria.

The sample groups must be assigned randomly.

There must be a viable control group.

Only one variable can be manipulated and tested. It is possible to test more than one,

but such experiments and their statistical analysis tend to be cumbersome and difficult.

The tested subjects must be randomly assigned to either control or experimental groups.

Advantages

The results of a true experimental design can be statistically analyzed and so there can be little

argument about the results.

It is also much easier for other researchers to replicate the experiment and validate the results.

For physical sciences working with mainly numerical data, it is much easier to manipulate one

variable, so true experimental design usually gives a yes or no answer.

Disadvantages

Whilst perfect in principle, there are a number of problems with this type of design. Firstly, they

can be almost too perfect, with the conditions being under complete control and not being

representative of real world conditions.

For psychologists and behavioral biologists, for example, there can never be any guarantee that

a human or living organism will exhibit ‘normal’ behavior under experimental conditions.

True experiments can be too accurate and it is very difficult to obtain a complete rejection or

acceptance of a hypothesis because the standards of proof required are so difficult to reach.

True experiments are also difficult and expensive to set up. They can also be very impractical.

While for some fields, like physics, there are not as many variables so the design is easy, for

social sciences and biological sciences, where variations are not so clearly defined it is much

more difficult to exclude other factors that may be affecting the manipulated variable.

Summary

True experimental design is an integral part of science, usually acting as a final test of a

hypothesis. Whilst they can be cumbersome and expensive to set up, literature reviews,

qualitative research and descriptive research can serve as a good precursor to generate a

testable hypothesis, saving time and money.

Whilst they can be a little artificial and restrictive, they are the only type of research that is

accepted by all disciplines as statistically provable.

Random Sampling Error

Random sampling errors are one type of experimental error that everybody should know.

Anyone who reads polls on the internet, or in newspapers, should be aware that sampling errors

could vastly influence the data and lead people to draw incorrect conclusions.

To further compound the random sampling errors, many survey companies, newspapers and

pundits are well aware of this, and deliberately manipulate polls to give favorable results.

In any experiment where it is impossible to sample an entire population, usually due to

practicality and expense, a representative sample must be used.

Of course, when you use a sample group, it can never fully match the entire population, and

there will always be some likelihood of random sampling error.

Any researcher must strive to ensure that the sample is as representative as possible, and

statistical tests have inbuilt checks and balances to take this into account.

To illustrate how to ensure that your statistics are as accurate as possible, we are going to use

the example of an opinion poll. These are one of the most commonly misinterpreted

representations of data, and failure to take into account the nuances of statistics can paint an

incorrect picture.

Margin of Error - A False Picture

The problem is, when you see an opinion poll in a newspaper or internet site, you will usually

see a margin of error, such a + or - 3%. The temptation is to think that the polls will be accurate

within this figure.

For example, if a poll gives one political party (A) a 42% share of the vote, and the other (B)

39%, this opens up a number of possible results. (A) could have 45%, (B) 36%. Both could be

39% or (B) could actually be ahead, 42% versus 39%. Of course, the results could show any

variation in between those extremes. Complicated enough?

To complicate the picture further, even this random sampling error can be wildly inaccurate. Any

opinion poll may give the margin of error, but this can convey a false sense of security and

make people assume that the results 'must' lie within this range.

In fact, these figures could actually be completely wrong, and the numbers are only ever an

estimate.

The Problem With Random Sampling Error

The problem is that these results only show the random sampling error within that specific

group. They show the chances of the results in that group occurring purely by chance, exactly

like the 95% confidence margin employed by many scientific researchers.

However, this is a very narrow definition and is often misunderstood.

In an opinion poll, there is no guarantee that the sample of 1000 or 10 000 people is truly

representative of the larger population as a whole.

There have been many extremely inaccurate polls conducted over the years, and they fell down

due to poor design and not understanding all of the relevant factors.

For example, an opinion poll company conducting telephone polls may make the mistake of only

telephoning during office hours, when most of the population is at work, skewing the data.

In addition, poorer families do not always have a fixed line telephone and use unregistered cell

phones, again leaving a huge potential for inaccuracy. The margins of error would be perfectly

acceptable, in these cases, but the overall findings would still be horribly wrong.

Modern polling companies are very skilled at designing polls to select samples from many

elements of the population, and via various media, so big errors rarely happen. Despite this,

opinion polls must always be taken as a guide only, not an exact representation of how an

election is likely to unfold.

Random Sampling Error and Experimental Design

The mistakes made by pollsters relate directly to any type of experiment involving random

sample groups.

Statistics can only work with the data provided and, if your design is poorly thought out, will not

be able to cover up these errors. Garbage in definitely equals garbage out.

Bibliography

Husch, B. (1971). Planning a Forest Inventory. Rome, Italy: Food and Agriculture Organization

of the United Nations

Urdan, T.C. (2005). Statistics in Plain English, Mahwah, NJ: Lawrence Erlbaum

Weisberg, H.F. (2005).The Total Survey Error Approach: A Guide to the New Science of Survey

Research. Chicago: University of Chicago Press

Scientists frequently use statistics to analyze their results. Why do researchers use statistics?

Statistics can help understand a phenomenon by confirming or rejecting a hypothesis. It is vital

to how we acquire knowledge to most scientific theories.

You don't need to be a scientist though; anyone wanting to learn about how researchers can get

help from statistics may want to read this statistics tutorial for the scientific method.

What is Statistics?

Research Data

This section of the statistics tutorial is about understanding how data is acquired and used.

The results of a science investigation often contain much more data or information than the

researcher needs. This data-material, or information, is called raw data.

To be able to analyze the data sensibly, the raw data is processed into "output data". There are

many methods to process the data, but basically the scientist organizes and summarizes the

raw data into a more sensible chunk of data. Any type of organized information may be called a

"data set".

Then, researchers may apply different statistical methods to analyze and understand the data

better (and more accurately). Depending on the research, the scientist may also want to use

statistics descriptively or for exploratory research.

What is great about raw data is that you can go back and check things if you suspect something

different is going on than you originally thought. This happens after you have analyzed the

meaning of the results.

The raw data can give you ideas for new hypotheses, since you get a better view of what is

going on. You can also control the variables which might influence the conclusion (e.g. third

variables). In statistics, a parameter is any numerical quantity that characterizes a given

population or some aspect of it.

Central Tendency and Normal Distribution

This part of the statistics tutorial will help you understand distribution, central tendency and how

it relates to data sets.

Much data from the real world is normal distributed, that is, a frequency curve, or a frequency

distribution, which has the most frequent number near the middle. Many experiments rely on

assumptions of a normal distribution. This is a reason why researchers very often measure the

central tendency in statistical research, such as the mean(arithmetic mean or geometric mean),

median or mode.

The central tendency may give a fairly good idea about the nature of the data (mean, median

and mode shows the "middle value"), especially when combined with measurements on how the

data is distributed. Scientists normally calculate the standard deviation to measure how the data

is distributed.

But there are various methods to measure how data is distributed: variance, standard deviation,

standard error of the mean, standard error of the estimate or "range" (which states the

extremities in the data).

To create the graph of the normal distribution for something, you'll normally use the arithmetic

mean of a "big enough sample" and you will have to calculate the standard deviation.

However, the sampling distribution will not be normally distributed if the distribution is skewed

(naturally) or has outliers (often rare outcomes or measurement errors) messing up the data.

One example of a distribution which is not normally distributed is the F-distribution, which is

skewed to the right.

So, often researchers double check that their results are normally distributed using range,

median and mode. If the distribution is not normally distributed, this will influence which

statistical test/method to choose for the analysis.

Other Tools

Quartile

Trimean

Hypothesis Testing - Statistics Tutorial

How do we know whether a hypothesis is correct or not?

Why use statistics to determine this?

Using statistics in research involves a lot more than make use of statistical formulas or getting to

know statistical software.

Making use of statistics in research basically involves

1. Learning basic statistics

2. Understanding the relationship between probability and statistics

3. Comprehension of the two major branches in statistics: descriptive statistics and

inferential statistics.

4. Knowledge of how statistics relates to the scientific method.

Statistics in research is not just about formulas and calculation. (Many wrong conclusions have

been conducted from not understanding basic statistical concepts)

Statistics inference helps us to draw conclusions from samples of a population.

When conducting experiments, a critical part is to test hypotheses against each other. Thus, it is

an important part of the statistics tutorial for the scientific method.

Hypothesis testing is conducted by formulating an alternative hypothesis which is tested against

the null hypothesis, the common view. The hypotheses are tested statistically against each

other.

The researcher can work out a confidence interval, which defines the limits when you will regard

a result as supporting the null hypothesis and when the alternative research hypothesis is

supported.

This means that not all differences between the experimental group and the control group can

be accepted as supporting the alternative hypothesis - the result need to differ significantly

statistically for the researcher to accept the alternative hypothesis. This is done using a

significance test (another article).

Caution though, data dredging, data snooping or fishing for data without later testing your

hypothesis in a controlled experiment may lead you to conclude on cause and effect even

though there is no relationship to the truth.

Depending on the hypothesis, you will have to choose between one-tailed and two tailed tests.

Sometimes the control group is replaced with experimental probability - often if the research

treats a phenomenon which is ethically problematic, economically too costly or overly time-

consuming, then the true experimental design is replaced by a quasi-experimental approach.

Often there is a publication bias when the researcher finds the alternative hypothesis correct,

rather than having a "null result", concluding that the null hypothesis provides the best

explanation.

If applied correctly, statistics can be used to understand cause and effect between research

variables.

It may also help identify third variables, although statistics can also be used to manipulate and

cover up third variables if the person presenting the numbers does not have honest intentions

(or sufficient knowledge) with their results.

Misuse of statistics is a common phenomenon, and will probably continue as long as people

have intentions about trying to influence others. Proper statistical treatment of experimental data

can thus help avoid unethical use of statistics. Philosophy of statistics involves justifying proper

use of statistics, ensuring statistical validity and establishing the ethics in statistics.

Here is another great statistics tutorial which integrates statistics and the scientific method.

Reliability and Experimental Error

Statistical tests make use of data from samples. These results are then generalized to the

general population. How can we know that it reflects the correct conclusion?

Contrary to what some might believe, errors in research are an essential part of significance

testing. Ironically, the possibility of a research error is what makes the research scientific in the

first place. If a hypothesis cannot be falsified (e.g. the hypothesis has circular logic), it is not

testable, and thus not scientific, by definition.

If a hypothesis is testable, to be open to the possibility of going wrong. Statistically this opens up

the possibility of getting experimental errors in your results due to random errors or other

problems with the research. Experimental errors may also be broken down into Type-I error and

Type-II error. ROC Curves are used to calculate sensitivity between true positives and false

positives.

A power analysis of a statistical test can determine how many samples a test will need to have

an acceptable p-value in order to reject a false null hypothesis.

The margin of error is related to the confidence interval and the relationship between statistical

significance, sample size and expected results. The effect size estimate the strength of the

relationship between two variables in a population. It may help determine the sample size

needed to generalize the results to the whole population.

Replicating the research of others is also essential to understand if the results of the research

were a result which can be generalized or just due to a random "outlier experiment". Replication

can help identify both random errors and systematic errors (test validity).

Cronbach's Alpha is used to measure the internal consistency or reliability of a test score.

Replicating the experiment/research ensures the reliability of the results statistically.

What you often see if the results have outliers, is a regression towards the mean, which then

makes the result not be statistically different between the experimental and control group.

Statistical Tests

Here we will introduce a few commonly used statistics tests/methods, often used by

researchers.

Relationship Between Variables

The relationship between variables is very important to scientists. This will help them to

understand the nature of what they are studying. A linear relationship is when two variables

varies proportionally, that is, if one variable goes up, the other variable will also go up with the

same ratio. A non-linear relationship is when variables do not vary proportionally. Correlation is

a a way to express relationship between two data sets or between two variables.

Measurement scales are used to classify, categorize and (if applicable) quantify variables.

Pearson correlation coefficient (or Pearson Product-Moment Correlation) will only express the

linear relationship between two variables. Spearman rho is mostly used for linear relationships

when dealing with ordinal variables. Kendall's tau (τ) coefficient can be used to measure

nonlinear relationships.

Partial Correlation (and Multiple Correlation) may be used when controlling for a third variable.

Predictions

The goal of predictions is to understand causes. Correlation does not necessarily mean

causation. With linear regression, you often measure a manipulated variable.

What is the difference between correlation and linear regression? Basically, a correlational

study looks at the strength between the variables whereas linear regression is about the best fit

line in a graph.

Regression analysis and other modeling tools

Linear Regression

Multiple Regression

A Path Analysis is an extension of the regression model

A Factor Analysis attempts to uncover underlying factors of something.

The Meta-Analysis frequently make use of effect size

Bayesian Probability is a way of predicting the likelihood of future events in an interactive way,

rather than to start measuring and then get results/predictions.

Testing Hypotheses Statistically

Student's t-test is a test which can indicate whether the null hypothesis is correct or not. In

research it is often used to test differences between two groups (e.g. between a control

group and an experimental group).

The t-test assumes that the data is more or less normally distributed and that the variance is

equal (this can be tested by the F-test).

Student's t-test:

Independent One-Sample T-Test

Independent Two-Sample T-Test

Dependent T-Test for Paired Samples

Wilcoxon Signed Rank Test may be used for non-parametric data.

A Z-Test is similar to a t-test, but will usually not be used on sample sizes below 30.

A Chi-Square can be used if the data is qualitative rather than quantitative.

Comparing More Than Two Groups

An ANOVA, or Analysis of Variance, is used when it is desirable to test whether there are

different variability between groups rather than different means. Analysis of Variance can also

be applied to more than two groups. The F-distribution can be used to calculate p-values for the

ANOVA.

Analysis of Variance

One way ANOVA

Two way ANOVA

Factorial ANOVA

Repeated Measures and ANOVA

Nonparametric Statistics

Some common methods using nonparametric statistics:

Cohen's Kappa

Mann-Whitney U-test

Spearman's Rank Correlation Coefficient

Other Important Terms in Statistics

Research Methodology

Key Concepts of the Scientific Method

There are several important aspects to research methodology. This is a summary of the key

concepts in scientific research and an attempt to erase some common misconceptions in

science.

Research Methodology

key Concepts of the Scientific Method

There are several important aspects to research methodology. This is a summary of the key

concepts in scientific research and an attempt to erase some common misconceptions in

science.

General Question

The starting point of most new research is to formulate a general question about an area of

research and begin the process of defining it.

This initial question can be very broad, as the later research, observation and narrowing down

will hone it into a testable hypothesis.

For example, a broad question might ask 'whether fish stocks in the North Atlantic are declining

or not', based upon general observations about smaller yields of fish across the whole area.

Reviewing previous research will allow a general overview and will help to establish a more

specialized area.

Unless you have an unlimited budget and huge teams of scientists, it is impossible to research

such a general field and it needs to be pared down. This is the method of trying to sample one

small piece of the whole picture and gradually contribute to the wider question.

Narrowing Down

The research stage, through a process of elimination, will narrow and focus the research area.

This will take into account budgetary restrictions, time, available technology and practicality,

leading to the proposal of a few realistic hypotheses.

Eventually, the researcher will arrive at one fundamental hypothesis around which the

experiment can be designed.

Designing the Experiment

This stage of the scientific method involves designing the steps that will test and evaluate the

hypothesis, manipulating one or more variables to generate analyzable data.

The experiment should be designed with later statistical tests in mind, by making sure that the

experiment has controls and a large enough sample group to provide statistically valid results.

Observation

This is the midpoint of the steps of the scientific method and involves observing and recording

the results of the research, gathering the findings into raw data.

The observation stage involves looking at what effect the manipulated variables have upon the

subject, and recording the results.

Analysis

The scope of the research begins to broaden again, as statistical analyses are performed on the

data, and it is organized into an understandable form.

The answers given by this step allow the further widening of the research, revealing some

trends and answers to the initial questions.

Conclusions and Publishing

This stage is where, technically, the hypothesis is stated as proved or disproved.

However, the bulk of research is never as clear-cut as that, and so it is necessary to filter the

results and state what happened and why. This stage is where interesting results can be

earmarked for further research and adaptation of the initial hypothesis.

Even if the hypothesis was incorrect, maybe the experiment had a flaw in its design or

implementation. There may be trends that, whilst not statistically significant, lead to further

research and refinement of the process.

The results are usually published and shared with the scientific community, allowing verification

of the findings and allowing others to continue research into other areas.

Cycles

This is not the final stage of the steps of the scientific method, as it generates data and ideas to

recycle into the first stage.

The initial and wider research area can again be addressed, with this research one of the many

individual pieces answering the whole question.

Building up understanding of a large area of research, by gradually building up a picture, is the

true path of scientific advancement. One great example is to look at the work of J J Thomson,

who gradually inched towards his ultimate answer.

Research Variables

The research variables, of any scientific experiment or research process, are factors that can be

manipulated and measured.

Any factor that can take on different values is a scientific variable and influences the outcome of

experimental research.

Gender, color and country are all perfectly acceptable variables, because they are inherently

changeable.

Most scientific experiments measure quantifiable factors, such as time or weight, but this is not

essential for a component to be classed as a variable.

As an example, most of us have filled in surveys where a researcher asks questions and asks

you to rate answers. These responses generally have a numerical range, from ‘1 - Strongly

Agree’ through to ‘5 - Strongly Disagree’. This type of measurement allows opinions to be

statistically analyzed and evaluated.

Dependent and Independent Variables

The key to designing any experiment is to look at what research variables could affect the

outcome.

There are many types of variable but the most important, for the vast majority of research

methods, are the independent and dependent variables.

A researcher must determine which variable needs to be manipulated to generate quantifiable

results.

The independent variable is the core of the experiment and is isolated and manipulated by the

researcher. The dependent variable is the measurable outcome of this manipulation, the results

of the experimental design. For many physical experiments, isolating the independent variable

and measuring the dependent is generally easy.

If you designed an experiment to determine how quickly a cup of coffee cools, the manipulated

independent variable is time and the dependent measured variable is temperature.

In other fields of science, the variables are often more difficult to determine and an experiment

needs a robust design. Operationalization is a useful tool to measure fuzzy concepts which do

not have one obvious variable.

The Difficulty of Isolating Variables

In biology, social science and geography, for example, isolating a single independent variable is

more difficult and any experimental design must consider this.

For example, in a social research setting, you might wish to compare the effect of different foods

upon hyperactivity in children. The initial research and inductive reasoning leads you to

postulate that certain foods and additives are a contributor to increased hyperactivity. You

decide to create a hypothesis and design an experiment, to establish if there is solid evidence

behind the claim.

The type of food is an independent variable, as is the amount eaten, the period of time

and the gender and age of the child. All of these factors must be accounted for during

the experimental design stage. Randomization and controls are generally used to

ensure that only one independent variable is manipulated.

To eradicate some of these research variables and isolate the process, it is essential to

use various scientific measurements to nullify or negate them.

For example, if you wanted to isolate the different types of food as the manipulated

variable, you should use children of the same age and gender.

The test groups should eat the same amount of the food at the same times and the

children should be randomly assigned to groups. This will minimize the physiological

differences between children. A control group, acting as a buffer against unknown

research variables, might involve some children eating a food type with no known links

to hyperactivity.

In this experiment, the dependent variable is the level of hyperactivity, with the resulting

statistical tests easily highlighting any correlation. Depending upon the results, you

could try to measure a different variable, such as gender, in a follow up experiment.

Converting Research Variables Into Constants

Ensuring that certain research variables are controlled increases the reliability and

validity of the experiment, by ensuring that other causal effects are eliminated. This

safeguard makes it easier for other researchers to repeat the experiment and

comprehensively test the results.

What you are trying to do, in your scientific design, is to change most of the variables

into constants, isolating the independent variable. Any scientific research does contain

an element of compromise and inbuilt error, but eliminating other variables will ensure

that the results are robust and valid.

Dependent Variable

Martyn Shuttleworth 64.9K reads

In any true experiment, a researcher manipulates an independent variable, to influence a

dependent variable, or variables.

A well-designed experiment normally incorporate one or two independent variables, with every

other possible factor eliminated, or controlled. There may be more than two dependent variables

in any experiment.

For example, a researcher might wish to establish the effect of temperature on the rate of plant

growth; temperature is the independent variable. They could regard growth as height, weight,

number of fruits produced, or all of these. A whole range of dependent variables arises from one

independent variable.

In any experimental design, the researcher must determine that there is a definite causal link

between the independent and dependent variable.

This reduces the risk of 'correlation and causation' errors. Controlled variables are used to

reduce the possibility of any other factor influencing changes in the dependent variable, known

as confounding variables.

In the above example, the plants must all be given the same amount of water, or this factor

could obscure any link between temperature and growth.

The relationship between the independent variable and dependent variable is the basis of most

statistical tests, which establish whether there is a real correlation between the two. The results

of these tests allow the researcher to accept or reject the null hypothesis, and draw conclusions.

Independent Variable

Martyn Shuttleworth 202.9K reads 1 Comment

The independent variable, also known as the manipulated variable, lies at the heart of any

quantitative experimental design.

This is the factor manipulated by the researcher, and it produces one or more results, known as

dependent variables. There are often not more than one or two independent variables tested in

an experiment, otherwise it is difficult to determine the influence of each upon the final results.

There may be more than several dependent variables, because manipulating the independent

can influence many different things.

For example, an experiment to test the effects of a certain fertilizer, upon plant growth, could

measure height, number of fruits and the average weight of the fruit produced. All of these are

valid analyzable factors, arising from the manipulation of one independent variable, the amount

of fertilizer.

Potential Complexities of the Independent Variable

The term independent variable is often a source of confusion; many people assume that the

name means that the variable is independent of any manipulation.

The name arises because the variable is isolated from any other factor, allowing experimental

manipulation to establish analyzable results.

Some research papers appear to give results manipulating more than one experimental

variable, but this is usually a false impression.

Each manipulated variable is likely to be an experiment in itself, one area where the words

'experiment' and 'research' differ. It is simply more convenient for the researcher to bundle them

into one paper, and discuss the overall results.

The botanical researcher above might also study the effects of temperature, or the amount of

water on growth, but these must be performed as discrete experiments, with only the conclusion

and discussion amalgamated at the end.

Independent Variables - Examples

As an example of an experiment with easily defined experimental variables, Mendel's famous

Pea Plant Experiment is a good choice.

The Austrian monk cross-pollinated pea plants, trying to establish which characteristics were

passed down through the generations. In this case, the inheritable characteristic of the parent

plant was the independent variable. For example, when plants with green seedpods were

crossed with plants with yellow seedpods, pod color was the independent variable.

In the Bandura Bobo Doll experiment, whether the children were exposed to an aggressive

adult, or to a passive adult, was the independent variable.

This experiment is a prime example of how the concept of experimental variables can become a

little complex. He also studied the differences between boys and girls, with gender as an

independent variable. Surely, this is breaking the rules of only having one manipulated variable!

In fact, this is a prime example of performing multiple experiments at the same time. If you study

carefully the structure of the research design, you will see that the Bobo Doll Experiment should

have been called the Bobo Doll Experiments.

It was actually four experiments, each with their own hypothesis and variables, running

concurrently. It would have been expensive, and possibly unethical, to test the children four

times and, if the same children were used each time, their behavior may have changed with

repetition.

Careful design allowed Bandura to test different hypotheses as part of the same research.

Statistically Significant Results

Statistically significant results are those that are interpreted not likely to have occurred purely by

chance and thereby have other underlying causes for their occurrence.

Whenever a statistical analysis is performed and results interpreted, there is always a finite

chance that the results are purely by chance. This is an inherent limitation of any statistical

analysis and cannot be done away with. Also, mistakes such as measurement errors may cause

the experimenter to misinterpret the results.

However, the probability that the process was simply a chance encounter can be calculated,

and a minimum threshold of statistical significance can be set. If the results are obtained such

that the probability that they are simply a chance process is less than this threshold of

significance, then we can say the results are not due to chance.

Common statistically significant levels are 5%, 1%, etc.

In terms of null hypothesis, the concept of statistical significance can be understood to be the

minimum level at which the null hypothesis can be rejected. This means if the experimenter sets

his statistical significance level at 5% and the probability that the results are a chance process is

3%, then the experimenter can claim that the null hypothesis can be rejected.

In this case, the experimenter will call his results to be statistically significant. Lower the

significance level, higher the confidence.

Statistically significant results are required for many practical cases of experimentation in

various branches of research. The choice of the statistical significance level is influenced by a

number of parameters and changes with different experiments.

In most cases of practical consideration, however, the distribution of parameters or qualities

follows a normal distribution, which is also the simplest case under consideration. However,

care should always be taken to account for other distributions within the given population.

While determining significant results statistically, it is important to note that it is impossible to

use statistics to prove that the difference in levels of two parameters is zero. This means that

the results of a significant analysis should not be interpreted as meaning there was no

difference. The only thing that the statistical analysis can state is that the experiment failed to

find any difference.

English

Home Research Methods

Controlled Variables

Controlled Variables

Martyn Shuttleworth 87.9K reads 1 Comment

Controlled variables are variables that is sometimes overlooked by researchers, but it is usually

far more important than the dependent or independent variables.

A failure to isolate the controlled variables, in any experimental design, will seriously

compromise the internal validity. This oversight may lead to confounding variables ruining the

experiment, wasting time and resources, and damaging the researcher's reputation.

In any experimental design, a researcher will be manipulating one variable, the independent

variable, and studying how that affects the dependent variables.

A failure to isolate the controlled variables will compromise the internal validity.

Most experimental designs measures only one or two variables at a time. Any other factor,

which could potentially influence the results, must be correctly controlled. Its effect upon the

results must be standardized, or eliminated, exerting the same influence upon the different

sample groups.

For example, if you were comparing cleaning products, the brand of cleaning product would be

the only independent variable measured. The level of dirt and soiling, the type of dirt or stain,

the temperature of the water and the time of the cleaning cycle are just some of the variables

that must be the same between experiments. Failure to standardize even one of these

controlled variables could cause a confounding variable and invalidate the results.

Control Groups

In many fields of science, especially biology and behavioral sciences, it is very difficult to ensure

complete control, as there is a lot of scope for small variations.

Biological processes are subject to natural fluctuations and chaotic rhythms. The key is to use

established operationalization techniques, such as randomization and double blind experiments.

These techniques will control and isolate these variables, as much as possible. If this proves

difficult, a control group is used, which will give a baseline measurement for the unknown

variables.

Sound statistical analysis will then eliminate these fluctuations from the results. Most statistical

tests have a certain error margin built in, and repetition and large sample groups will eradicate

the unknown variables.

There still needs to be constant monitoring and checks, but due diligence will ensure that the

experiment is as accurate as is possible.

The Value of Consistency

Controlled variables are often referred to as constants, or constant variables.

It is important to ensure that all these possible variables are isolated, because a type III error

may occur if an unknown factor influences the dependent variable. This is where the null

hypothesis is correctly rejected, but for the wrong reason.

In addition, inadequate monitoring of controlled variables is one of the most common causes of

researchers wrongly assuming that a correlation leads to causality.

Controlled variables are the road to failure in an experimental design, if not identified and

eliminated. Designing the experiment with controls in mind is often more crucial than

determining the independent variable.

Poor controls can lead to confounding variables, and will damage the internal validity of the

experiment.

Operationalization

Operationalization is the process of strictly defining variables into measurable factors. The

process defines fuzzy concepts and allows them to be measured, empirically and quantitatively.

For experimental research, where interval or ratio measurements are used, the scales are

usually well defined and strict.

Operationalization also sets down exact definitions of each variable, increasing the quality of the

results, and improving the robustness of the design.

For many fields, such as social science, which often use ordinal measurements,

operationalization is essential. It determines how the researchers are going to measure an

emotion or concept, such as the level of distress or aggression.

Such measurements are arbitrary, but allow others to replicate the research, as well as perform

statistical analysis of the results.

Fuzzy Concepts

Fuzzy concepts are vague ideas, concepts that lack clarity or are only partially true. These are

often referred to as "conceptual variables".

It is important to define the variables to facilitate accurate replication of the research process.

For example, a scientist might propose the hypothesis:

“Children grow more quickly if they eat vegetables.”

What does the statement mean by 'children'? Are they from America or Africa. What age are

they? Are the children boys or girls? There are billions of children in the world, so how do you

define the sample groups?

How is 'growth' defined? Is it weight, height, mental growth or strength? The statement does not

strictly define the measurable, dependent variable.

What does the term 'more quickly' mean? What units, and what timescale, will be used to

measure this? A short-term experiment, lasting one month, may give wildly different results than

a longer-term study.

The frequency of sampling is important for operationalization, too.

If you were conducting the experiment over one year, it would not be practical to test the weight

every 5 minutes, or even every month. The first is impractical, and the latter will not generate

enough analyzable data points.

What are 'vegetables'? There are hundreds of different types of vegetable, each containing

different levels of vitamins and minerals. Are the children fed raw vegetables, or are they

cooked? How does the researcher standardize diets, and ensure that the children eat their

greens?

Operationalization

The above hypothesis is not a bad statement, but it needs clarifying and strengthening, a

process called operationalization.

The researcher could narrow down the range of children, by specifying age, sex, nationality, or

a combination of attributes. As long as the sample group is representative of the wider group,

then the statement is more clearly defined.

Growth may be defined as height or weight. The researcher must select a definable and

measurable variable, which will form part of the research problem and hypothesis.

Again, 'more quickly' would be redefined as a period of time, and stipulate the frequency of

sampling. The initial research design could specify three months or one year, giving a

reasonable time scale and taking into account time and budget restraints.

Each sample group could be fed the same diet, or different combinations of vegetables. The

researcher might decide that the hypothesis could revolve around vitamin C intake, so the

vegetables could be analyzed for the average vitamin content.

Alternatively, a researcher might decide to use an ordinal scale of measurement, asking

subjects to fill in a questionnaire about their dietary habits.

Already, the fuzzy concept has undergone a period of operationalization, and the hypothesis

takes on a testable format.

The Importance of Operationalization

Of course, strictly speaking, concepts such as seconds, kilograms and centigrade are artificial

constructs, a way in which we define variables.

Pounds and Fahrenheit are no less accurate, but were jettisoned in favor of the metric system.

A researcher must justify their scale of scientific measurement.

Operationalization defines the exact measuring method used, and allows other scientists to

follow exactly the same methodology. One example of the dangers of non-operationalization is

the failure of the Mars Climate Orbiter.

This expensive satellite was lost, somewhere above Mars, and the mission completely failed.

Subsequent investigation found that the engineers at the sub-contractor, Lockheed, had used

imperial units instead of metric units of force.

A failure in operationalization meant that the units used during the construction and simulations

were not standardized. The US engineers used pound force, the other engineers and software

designers, correctly, used metric Newtons.

This led to a huge error in the thrust calculations, and the spacecraft ended up in a lower orbit

around Mars, burning up from atmospheric friction. This failure in operationalization cost

hundreds of millions of dollars, and years of planning and construction were wasted.

Conceptual Variables

Explorable.com 34.1K reads

Conceptual variables are often expressed in general, theoretical, qualitative, or subjective terms

and important in hypothesis building process.

Two levels of abstraction exist for our research activities and our understanding of research

outcomes. Everyone understands at conceptual level. For example, if you say "Computer

games sharpen children's minds" expresses a belief about a causal relationship at a conceptual

level. At this level of abstraction, the variables are called constructs or conceptual variables.

Constructs are the mental definitions of properties of events of objects that can vary. Definitions

of computer games and mental sharpness are examples of such constructs.

Now, computer games and mental sharpness need be defined and explained. It is important to

note that the empirical research activities are carried out at an operational level of abstraction

and empirical research acquire scores from cases on measures. These measures represent

operational variables. The variables can be made operational by the measures used to acquire

scores from the cases studied. For example, a question that asks children how many hours a

day they play computer games is an operational measure children's interest in computer games.

Conceptual variables are often expressed in general, theoretical, subjective, or qualitative

terms. The research hypothesis is usually starts at this level, for example. "Effect of nicotine

patch is poorer among people lacking mental determination to quit smoking".

To measure conceptual variables, an objective definition is often required. This may involve

having an easily available validated instrument, inferring an operational variable from theory,

establishing consensus or all three. In above example, we need to have a definition of effect of

nicotine patch and mental determination.

During this process, one needs to decide on measurement scale. The researcher may decide to

make effect of nicotine patch: yes/no" (nominal), or "none/low/moderate/high (ordinal) based on

definition of potency of a patch. For mental determination to quit smoking, you may need to do

the same: present/absent, or, more likely, use some ordinal scale based on a predesigned

questionnaire or third party evaluation.

Another example: if this is stated that

"The recovery in diabetic patient was quick among those patients without concurrent

cardiovascular problems"

Now, the recovery needs to be converted into some measureable variable e.g.

"maintenance of glucose levels over one year (continuous scale), as does cardiovascular

problems, e.g."

No history of previous heart attack, normal findings of ECG/Echocardiography/Color Doppler

and cardiac enzymes etc for evaluation of cardiovascular status (continuous scale).

English

Home

Research

Experiments

Experimental Research

Oskar Blakstad 1 read 15 Comments

Experimental research is commonly used in sciences such as sociology and

psychology, physics, chemistry, biology and medicine etc.

It is a collection of research designs which use manipulation and controlled

testing to understand causal processes. Generally, one or more variables

are manipulated to determine their effect on a dependent variable.

The experimental method

is a systematic and scientific approach to research in which the researcher manipulates one or

more variables, and controls and measures any change in other variables.

Experimental Research is often used where:

1. There is time priority in a causal relationship (cause precedes effect)

2. There is consistency in a causal relationship (a cause will always lead to the same

effect)

3. The magnitude of the correlation is great.

(Reference: en.wikipedia.org)

The word experimental research has a range of definitions. In the strict sense, experimental

research is what we call a true experiment.

This is an experiment where the researcher manipulates one variable, and control/randomizes

the rest of the variables. It has a control group, the subjects have been randomly assigned

between the groups, and the researcher only tests one effect at a time. It is also important to

know what variable(s) you want to test and measure.

A very wide definition of experimental research, or a quasi experiment, is research where the

scientist actively influences something to observe the consequences. Most experiments tend to

fall in between the strict and the wide definition.

A rule of thumb is that physical sciences, such as physics, chemistry and geology tend to define

experiments more narrowly than social sciences, such as sociology and psychology, which

conduct experiments closer to the wider definition.

Aims of Experimental Research

Experiments are conducted to be able to predict phenomenons. Typically, an experiment is

constructed to be able to explain some kind of causation. Experimental research is important to

society - it helps us to improve our everyday lives.

Identifying the Research Problem

After deciding the topic of interest, the researcher tries to define the research problem. This

helps the researcher to focus on a more narrow research area to be able to study it

appropriately. Defining the research problem helps you to formulate a research hypothesis,

which is tested against the null hypothesis.

The research problem is often operationalizationed, to define how to measure the research

problem. The results will depend on the exact measurements that the researcher chooses and

may be operationalized differently in another study to test the main conclusions of the study.

An ad hoc analysis is a hypothesis invented after testing is done, to try to explain why the

contrary evidence. A poor ad hoc analysis may be seen as the researcher's inability to accept

that his/her hypothesis is wrong, while a great ad hoc analysis may lead to more testing and

possibly a significant discovery.

Constructing the Experiment

There are various aspects to remember when constructing an experiment. Planning ahead

ensures that the experiment is carried out properly and that the results reflect the real world, in

the best possible way.

Sampling Groups to Study

Sampling groups correctly is especially important when we have more than one condition in the

experiment. One sample group often serves as a control group, whilst others are tested under

the experimental conditions.

Deciding the sample groups can be done in using many different sampling techniques.

Population sampling may chosen by a number of methods, such as randomization, "quasi-

randomization" and pairing.

Reducing sampling errors is vital for getting valid results from experiments. Researchers often

adjust the sample size to minimize chances of random errors.

Here are some common sampling techniques:

probability sampling

non-probability sampling

simple random sampling

convenience sampling

stratified sampling

systematic sampling

cluster sampling

sequential sampling

disproportional sampling

judgmental sampling

snowball sampling

quota sampling

Creating the Design

The research design is chosen based on a range of factors. Important factors when choosing

the design are feasibility, time, cost, ethics, measurement problems and what you would like to

test. The design of the experiment is critical for the validity of the results.

Typical Designs and Features in Experimental Design

Pretest-Posttest Design

Check whether the groups are different before the manipulation starts and the effect of

the manipulation. Pretests sometimes influence the effect.

Control Group

Control groups are designed to measure research bias and measurement effects, such

as the Hawthorne Effect or the Placebo Effect. A control group is a group not receiving

the same manipulation as the experimental group. Experiments frequently have 2

conditions, but rarely more than 3 conditions at the same time.

Randomized Controlled Trials

Randomized Sampling, comparison between an Experimental Group and a Control

Group and strict control/randomization of all other variables

Solomon Four-Group Design

With two control groups and two experimental groups. Half the groups have a pretest

and half do not have a pretest. This to test both the effect itself and the effect of the

pretest.

Between Subjects Design

Grouping Participants to Different Conditions

Within Subject Design

Participants Take Part in the Different Conditions - See also: Repeated Measures

Design

Counterbalanced Measures Design

Testing the effect of the order of treatments when no control group is available/ethical

Matched Subjects Design

Matching Participants to Create Similar Experimental- and Control-Groups

Double-Blind Experiment

Neither the researcher, nor the participants, know which is the control group. The results

can be affected if the researcher or participants know this.

Bayesian Probability

Using bayesian probability to "interact" with participants is a more "advanced"

experimental design. It can be used for settings were there are many variables which are

hard to isolate. The researcher starts with a set of initial beliefs, and tries to adjust them

to how participants have responded

Pilot Study

It may be wise to first conduct a pilot-study or two before you do the real experiment. This

ensures that the experiment measures what it should, and that everything is set up right.

Minor errors, which could potentially destroy the experiment, are often found during this

process. With a pilot study, you can get information about errors and problems, and improve the

design, before putting a lot of effort into the real experiment.

If the experiments involve humans, a common strategy is to first have a pilot study with

someone involved in the research, but not too closely, and then arrange a pilot with a person

who resembles the subject(s). Those two different pilots are likely to give the researcher good

information about any problems in the experiment.

Conducting the Experiment

An experiment is typically carried out by manipulating a variable, called the independent

variable, affecting the experimental group. The effect that the researcher is interested in, the

dependent variable(s), is measured.

Identifying and controlling non-experimental factors which the researcher does not want to

influence the effects, is crucial to drawing a valid conclusion. This is often done by controlling

variables, if possible, or randomizing variables to minimize effects that can be traced back to

third variables. Researchers only want to measure the effect of the independent variable(s)

when conducting an experiment, allowing them to conclude that this was the reason for the

effect.

Analysis and Conclusions

In quantitative research, the amount of data measured can be enormous. Data not prepared to

be analyzed is called "raw data". The raw data is often summarized as something called "output

data", which typically consists of one line per subject (or item). A cell of the output data is, for

example, an average of an effect in many trials for a subject. The output data is used for

statistical analysis, e.g. significance tests, to see if there really is an effect.

The aim of an analysis is to draw a conclusion, together with other observations. The researcher

might generalize the results to a wider phenomenon, if there is no indication of confounding

variables "polluting" the results.

If the researcher suspects that the effect stems from a different variable than the independent

variable, further investigation is needed to gauge the validity of the results. An experiment is

often conducted because the scientist wants to know if the independent variable is having any

effect upon the dependent variable. Variables correlating are not proof that there is causation.

Experiments are more often of quantitative nature than qualitative nature, although it happens.

Examples of Experiments

This website contains many examples of experiments. Some are not true experiments, but

involve some kind of manipulation to investigate a phenomenon. Others fulfill most or all criteria

of true experiments.

Here are some examples of scientific experiments:

Social Psychology

Stanley Milgram Experiment - Will people obey orders, even if clearly dangerous?

Asch Experiment - Will people conform to group behavior?

Stanford Prison Experiment - How do people react to roles? Will you behave differently?

Good Samaritan Experiment - Would You Help a Stranger? - Explaining Helping

Behavior

Genetics

Law Of Segregation - The Mendel Pea Plant Experiment

Transforming Principle - Griffith's Experiment about Genetics

Defining a Research Problem

Martyn Shuttleworth 382.4K reads 13 Comments

Defining a research problem is the fuel that drives the scientific process, and is the foundation of

any research method and experimental design, from true experiment to case study.

It is one of the first statements made in any research paper and, as well as defining the

research area, should include a quick synopsis of how the hypothesis was arrived at.

Operationalization is then used to give some indication of the exact definitions of the variables,

and the type of scientific measurements used.

This will lead to the proposal of a viable hypothesis. As an aside, when scientists are putting

forward proposals for research funds, the quality of their research problem often makes the

difference between success and failure.

Structuring the Research Problem

Look at any scientific paper, and you will see the research problem, written almost like a

statement of intent.

Defining a research problem is crucial in defining the quality of the answers, and determines the

exact research method used. A quantitative experimental design uses deductive reasoning to

arrive at a testable hypothesis.

Qualitative research designs use inductive reasoning to propose a research statement.

Defining a Research Problem

Formulating the research problem begins during the first steps of the scientific process.

As an example, a literature review and a study of previous experiments, and research, might

throw up some vague areas of interest.

Many scientific researchers look at an area where a previous researcher generated some

interesting results, but never followed up. It could be an interesting area of research, which

nobody else has fully explored.

A scientist may even review a successful experiment, disagree with the results, the tests used,

or the methodology, and decide to refine the research process, retesting the hypothesis.

This is called the conceptual definition, and is an overall view of the problem. A science report

will generally begin with an overview of the previous research and real-world observations. The

researcher will then state how this led to defining a research problem.

The Operational Definitions

The operational definition is the determining the scalar properties of the variables.

For example, temperature, weight and time are usually well known and defined, with only the

exact scale used needing definition. If a researcher is measuring abstract concepts, such as

intelligence, emotions, and subjective responses, then a system of measuring numerically

needs to be established, allowing statistical analysis and replication.

For example, intelligence may be measured with IQ and human responses could be measured

with a questionnaire from ‘1- strongly disagree’, to ‘5 - strongly agree’.

Behavioral biologists and social scientists might design an ordinal scale for measuring and

rating behavior. These measurements are always subjective, but allow statistics and replication

of the whole research method. This is all an essential part of defining a research problem.

Examples of Defining a Research Problem

An anthropologist might find references to a relatively unknown tribe in Papua New Guinea.

Through inductive reasoning, she arrives at the research problem and asks,

‘How do these people live and how does their culture relate to nearby tribes?’

She has found a gap in knowledge, and she seeks to fill it, using a qualitative case study,

without a hypothesis.

The Bandura Bobo Doll Experiment is a good example of using deductive reasoning to arrive at

a research problem and hypothesis.

Anecdotal evidence showed that violent behavior amongst children was increasing. Bandura

believed that higher levels of violent adult role models on television, was a contributor to this

rise. This was expanded into a hypothesis, and operationalization of the variables, and scientific

measurement scale, led to a robust experimental design.