Today, we are living in the info Era. We make many of our decisions, whether we intend to venture out to sea to fish, buy a new computer, invest in projects, built a new holiday resort, or even go to warfare, predicated on information that we gather. The more info we obtain, how fast we get them and how relevant they are really will have an effect on our decisions. However, more important than velocity or amount of information is whether the information we acquired is real or shows the truth or has been interpreted correctly. Unfortunatley, for various reasons, there are numerous information out there that is fake, half-truths, misinterpreted or maybe constructed, either intentionally or unintentionally. So how do we realize a certain information that people obtain is the truth? Is it the reality because Mr. X said so? Can we trust his words? Who is this Mr. X? Can we believe that him because he is the Prime Minister or Chief executive of the United States? How have he obtain this information in the first place? Has he got any ulterior purpose nourishing you with this information? So we commence to mistrust. But if we will suspect every information that comes, then we will have a significant problem making our daily decisions.

## Science

There is a dependence on some device where information generated from that system gets the highest probability of being true. This device is called Research. Science comes from the Latin phrase "scientia" this means "knowledge". So technology is something or mechanism of aquiring knowledge and is aimed at locating the truth. Scientists are in the business of making new knowledge and it is important that the new knowledge refect what's true. That is why the methodical community demands that scientists must have a high level of integrity and actually so that results using their research reflects the reality based on the reality gathered. If fake information were allowed to be diseminated, with time, nobody will have confidence in information made by the scientific community and that will be the finish of science. To prevent this from going on, a set of guidelines were put in location to be accompanied by experts in their acquisition of knowledge. It really is thus very very important to for young scientists to follow the Scientific Method in their research investigations.

As researchers, we also have to think medically. Our forces of reasoning must lead successfully to the most logical answers and reach reliable conclusions. Scientific thinking is based on three things i. e. the use of empirical research, practice logical reasoning and possessing a skeptical frame of mind. Empirical research is evidence that one may see, notice, touch, taste, or smell. It really is facts that others, besides yourself, can experience, which is repeatable. Empirical data is the only type of proof used by researchers to make decisions and reach sound conclusions. Logic is not an ability that we are blessed with. It is a skill or discipline that must be learned. Emotional, hopeful, and wishful thinking is more common than logical thinking because they are easier plus more cogenial to human mother nature. Most individuals would prefer to believe something is true because they feel, wish, or wish it were true, rather than deny their thoughts and allow that their beliefs are false. Having a Skeptical Frame of mind is to constantly question your beliefs and conclusions. Good experts constantly examine the evidence, arguments and reasons for their values. A skeptic supports values only tentatively, and can willingly dispose of them if new proof can prove often. We must own an open brain.

## Scientific Method

Science is about discovering reliable understanding of dynamics. Reliable knowledge is knowledge that has a high probability to be true because its veracity has been justified by a trusted method. The Scientific Method is an activity for assessing knowledge to explain observable events in aspect by natural triggers without assuming the living of the supernatural. Scientists use observations and reasoning to propose tentative explanations for natural phenomena, termed hypotheses. Predictions from these hypotheses are then analyzed by experiments, that ought to be reproducible. A significant facet of a hypothesis is the fact that it must be falsifiable, i. e. it must be conceivable to demonstrate the hypothesis to be false. Once a hypothesis is consistently verified through experiment, it is known as to be always a theory and new predictions are founded upon it. Scientific methods are means employed by scientific communities for building supportable, evidence-based understandings of our natural world.

There are four essential elements in just a medical method :

Characterizations (quantifications, observations and measurements)

Hypotheses (theoretical, hypothetical explanations of observations and measurements)

Predictions (reasoning including logical deduction from hypotheses and theories)

Experiments (tests out of all the above)

A pragmatic plan of the four above points is sometimes offered as a guide for proceeding:

Define the question

Gather literature, information and resources

Form your hypothesis

Plan the experiment

Do the experiment and collect data

Analyze the noticed data

Interpret data and attract conclusions that provide as a starting place for new hypotheses

Communicate your results

## Statistical Analysis

A very important element of the Scientific Method is the statistical analysis of your accumulated data or observations. The method that you analyse the info, whether done appropriately or improperly, will ultimately determine the conclusions from your research. Any body who must collect data, prepare reports, read studies and draw clever conclusions from them must have a good knowledge of statistics. You can find universal approval of figures as an important tool for all types of research. It has also led to a rise in the number and diversity of statistical techniques. Although this variety indicates the option of appropriate statistical techniques for most research problems, it also signifies the difficulty of complementing the best technique to a specific test. Choosing the correct statistical procedure for a given test must be predicated on expertise in information and in the topic matter under study. Information, like any useful tool, can be misused either intentionally or by well-meaning analysts who know too little about research or statistical principles and steps.

## Why do we are in need of Statistics?

Diversity is an complex property of aspect. It is with variety that life on earth can persist as it allows evolution and adaptation to the ever changing environment on the planet. With variety, there exist variation. Variation occurs just about everywhere and in almost anything. There is variance in height, weight, shade, smell, etc. Thus for each and every population, there is certainly variant in physical, chemical substance and natural properties. As such, before we can say that there surely is a difference in a specific parameter between two population, we must take into consideration this variation. We have to show establish that even with the variation which exist within each population for the parameter in question, it continues to be highly possible that differences exist between your two populations. Statistical techniques were developed to do that. To take into account the variations before making a decision whether we can securely say that both populations will vary. When there is no variation, you will see no dependence on statistics.

## Types of Statistics in Marine Science Research

There are basically two types of statistics

## a) Descriptive reports.

Reduction of large public of uncooked data to a manageable form - e. g. graphs, furniture, steps of central inclination and procedures of dispersion.

## b) Predictive figures.

The data we collect is nearly always a sample of all the data we could have collected, and we want to put it to use to bring conclusions about the complete population. The capability to make such generalised conclusions, inferring characteristics of the complete from characteristics of the sample sits within the realm of inferential or predictive statistics.

In Predictive Statistics, statistical examination are usually conducted on the sampled facts or data from which conclusions about the populace is drawn. The statistical examination usually starts with a hypothesis and based on the data in the data, the likelihood of a certain result of the hypothesis is determined.

## Hypothesis Testing

Hypothesis Tests is a way by which can help us make decisions concerning differences. It is a process of infering from an example or samples if to simply accept a certain affirmation about the populace. The affirmation itself is called the hypothesis. The hypothesis is examined based on evidence contained in the sample or examples. The hypothesis ought to be the simplest one possible with minimal number of unidentified factors. It is a prerequisite to the application of a statistical test.

## General technique in statistical hypothesis evaluation.

## a) Designate a nul hypothesis (H0).

The hypothesis of no difference. The hypothesis that nothing unusual has occurred or what is likely to happen according to some standard theory.

## b) Specify the alternate hypothesis (H1).

Example:

H0: There is no difference in expansion of fishes given with diet A and diet B.

H1: There is a difference in progress of fishes fed with diet A and diet B.

H0: The populace sampled conforms to the Normal Distribution.

H1: The populace sampled will not conform to the standard Distribution.

H0: Both samples participate in the same society.

H1: The two samples result from different populations.

## c) Check data.

From the data, see which of H0 or H1 is accurate.

The answer will either be

i) Not obvious

ii) Obvious

iii) Very obvious

Only in case i) can you go to do a statistical test. It is neither necessary or useful to do a great deal of arithmetic showing something that was apparent before you started. Statistics is not really a substitute for good sense.

## d) Specify the amount of significance, .

Specify the critical possibility level below which H0 will be turned down. It really is conventionally taken to be 0. 05 or 5% degree of significance (or 95% self-confidence boundaries) in biological statistics.

In statistics, we live testing for distinctions. We first expect that there surely is no difference, H0. Then we test for difference, H1.

Hence, the amount of significance is the maximum possibility of rejecting a genuine null hypothesis ( 5% level of rejecting H0 ) when it's actually right.

= probability of committing a sort I error (i. e. probability of rejecting H0 when it's actually right).

= probability of committing a Type II problem (i. e. probability of accepting H0 when it's actually not appropriate).

Null Hypothesis (H0)

## TRUE

## FALSE

## REJECT

Type I Error

Correct

## ACCEPT

Correct

Type II Error

It is better to commit a sort II error than a Type I error.

We won't know if we've committed a sort I error but the likelihood of committing it is given as < 5% (utilizing a =0. 05) or < 1% (using a = 0. 01).

or,

What is the likelihood, p, of earning the problem of rejecting Ho when Ho is in fact true ?

If p is very low then we reject Ho.

If p is high then we had better allow Ho.

How low should p be before we reject Ho ? depends upon the amount of significance, a, arranged by us (usually 0. 05).

## e) Calculate the likelihood, p.

Assuming that Ho is correct, calculate the likelihood, p, (using appropriate figures) of obtaining results as extreme, or even more extreme, than those discovered.

There are several statistical checks available. To be able to choose, we consider several properties of the various tests e. g.

i) Will be the assumptions of these assessments valid assumptions in my experiment ? Criticisms with an experiment is often highest scheduled to lack of consideration of the assumptions.

ii) The test should be neutral and consistent.

iii) The test should be more efficient in a few sense than the other tests.

## f) Comment.

We rarely have enough training or knowledge to carefully understand all the possible violations of assumptions inherent in the design and analysis of their research, although they are most surely aware of the hypothesis they want to test.

## Types of Statistical Tests

Various types of statistical testing can be found. However, we can generally divide them into Parametric and Non-parametric checks.

## a) Parametric test

For making inferences about population parameters by examining sample figures. Assumes that the adjustable in question practices (at least approximates) the normal distribution. For interval and ratio level data.

## b) Data transformation

Generally to normalise data which do not gratify the above mentioned assumption so that they may be analysed using parametric methods.

## c) Nonparametric test

To draw inferences about population, not parameters. Usually do not require knowlegde about people distribution (circulation free figures). Fast with less arithmetic but less powerful than parametric exams. For norminal and ordinal scale data. Note that interval and ratio scale data can be changed into ordinal data by ranking.

## Examples of parametric tests

## a) Testing distinctions between two means.

## 1) Z-test

Where society variance, S, is well known.

## 2) Student's t-test (One and two samples)

Where the estimate s can be used.

## 3) Paired test t-test

For paired examples.

## b) Testing variations between a set of sample means.

## 1) One-way ANOVA.

## 2) Two-way ANOVA with and without replications.

## 3) Multi-way ANOVA.

## 4) Latin-Square.

## 5) Multiple comparisons.

## a. A minimum of Significant difference, LSD.

## b. Tukey Test.

## c. Student-Newman-Keuls Test.

## d. New Duncan's Multiple Range Test.

## e. Style comparisons

## c) Testing differences between variances.

## 1) F-test

## 2) Bartlett's test

## d) Relationship and regression analysis

## Examples of Non-parametric tests

## a) Runs test

Test for randomness in a linear collection of nominal data.

## b) Chi-square Goodness-of-fit test

Test or compare observed frequency circulation with predicted/theoretical rate of recurrence distribution.

## c) Homogeneity Chi-square test and Contingency tables

Test or compare 2 detected occurrence distributions.

## d) Kolmogorov-Smirnov test

Goodness-of-fit test for ordinal size data. Uses cumulative consistency data alternatively than Chi-square. Powerful where n is small, Fi is small.

## e) Mann-Whitney U-test

Nonparametric treatment anologous to 2 sample students t-test.

## f) Wilcoxon's paired test test

Nonparametric procedure anologous to matched sample t-test.

## g) McNemar's test

Paired sample evaluation of nominal data.

## h) Kruskal-Wallis test

Nonparametric One-way ANOVA by rates.

## i) Freidman's test

Nonparametric randomised block design by ranks.

## j) Spearman's Ranking Correlation

Nonparametric correlation on ordinal data.

## Multivariate Statistics

Most of the Statistical methods mentioned previously are termed as Univariate information because they verify only one variable while the other are cared for as treatment groups of factors. However, there is an increasing use of Multivariate Analysis where the procedure will examine a number of variables at once largely to find patterns, human relationships and interactions between them. A few of the most widely used multivariate steps include:

a) Multiple regression and correlation.

Where one would like to establish maximal linear romantic relationships among three or more sets of parameters.

b) Principal Component Research.

To reduce the dimensionality of the initial data while reducing loss of information and identifying those that take into account almost all of the deviation in the population.

c) Factor Analysis.

Resolve the intercorrelations among factors into their fundamental causes.

d) Multivariate analysis of variance.

To determine whether the samples might have been drawn from an individual statistical population.

e) Discrimant Research.

To sort out the objects into their appropriate populations with minimal error.

f) Cluster evaluation.

To sort recently unpartitioned heterogeneous collection of objects into a series of collections and determine the relationship ships between your sets.