Understanding Correlation and its Types in Statistics

This article explains the concept of correlation in statistics and its different types, including positive, negative, and minimal correlation. It also discusses how correlation tools can be used to identify variables and apply linear regression to obtain the rate of change of one variable as a result of change of the other variable.

Correlation is a statistical instrument that characterizes the relationship between two
variables. It gives an understanding with regards to the course and force of relationship
between two variables. (Marc De Hert, 2011)
Two variables are said to be positively correlated when they move together a similar way.
This means that:
i. As one variable increase, so does the other one.
ii. As one variable decrease, so does the other one.
Example
 As the temperature goes up, ice cream sales also go up.
 When enrollment at college decreases, the number of teachers decreases.
Two factors are negatively correlated in the event that they move in inverse ways.
This means that:
i. An increase of one variable corresponds to decrease of the other
ii. Decrease of one variable corresponds to increase of the other
Example
 As More time is spent at work, less time is spent at home. As you spend more and
more time at your workplace, it becomes difficult to spend that much time at home
Consequently, the more time you spend at office, the lesser time you give to your
family.
 Higher expense ratio, less returns. If you have made investment in mutual funds, you
may have noticed that the higher your expense ratio is, the lesser are your investment
returns.
A minimal correlation is where there's no connection, or interdependence between the two
variables.
Example
 Eating Cheetos does not make you better at speaking French. They are not related,
there is no correlation
 Apple pie sales in Iowa are not related to the shoe sizes of third graders in the Ukraine

For the first case; There is positive correlation between the two variables because of the
impact of the playground which is connected directly to existence of basketball teams. This is
a long-term objective and outcomes that will see more college basketball and NBA teams
joining the league.
For the second case; The correlation between the two variables is minimal on account of the
non-linear relationship. The fact that we have high demographic of younger target market
would in turn result to development of more sporting facility; which is not the case here
thereby justifying the non-linear relationship. This is short term objective and outcomes
which will change over time because there will be great demand of development of more
sporting facility due to high demographic of younger target market.
For the third case; Positive correlation is observed between the two variables because of the
fact that the extreme warmth has an adverse effect to the health due to high temperatures
which makes the participants suffer greatly and thereby much preference is put on indoor
sporting facilities. Therefore, as the extreme warmth increases, high indoor sporting facilities
increases justifying the positive correlation between the two variables. This is long term
objective and outcome.
For the Fourth case; The two variables are negatively correlated on account that low-
income families live in environments of extreme, concentrated poverty while high-income
populations are shifting toward the suburbs. Low population density, low consumption rate
and the low settlement of low-income families are negatively correlated to high income
geographic area with high population density, high settlement and high consumption rate.
This makes the two variables move in opposite direction thereby justifying the negative
correlation between the two variables. This is a long-term objective and outcome.
Outdoor sports would grow tremendously and the trend will rise over subsequent years as the
implication of Big D incorporated.
Entrance into the indoor sporting products would see a fascinating advancement regarding the
market for the related items. The running business sector is rising just marginally; however, it
has just achieved a high and sound dimension (Li, 2011).
Correlation tools can be used to identify variables by applying simple linear regression.
However, we need to know independent and dependent variable. Independent variable is
variable that is steady and unaffected by alternate factors we are attempting to quantify.

Dependent variable is the variable that relies upon different elements that are estimated.
(MAKSIMENKO L.L., 2010)
Therefore, these variables are applied in a linear regression to obtain the rate of change of one
variable as a result of change of the other variable (DOGADOVA T.V., 2014)

References
DOGADOVA T.V., V. V. (2014). GUARANTEED PARAMETER ESTIMATION OF STOCHASTIC LINEAR
REGRESSION BY SAMPLE OF FIXED SIZE. 14.
Li, C. H. (2011). IEEE 2011 International Conference on Future Computer Science and Education
(ICFCSE) - Xi'an, China (2011.08.20-2011.08.21)] 2011 International Conference on Future
Computer Science and Education - Comparative Study of Sports Statistics Textbooks and
Othe. 4.
MAKSIMENKO L.L., B. I. (2010). THE MEDICAL STATISTICS. STATISTICAL METHODS IN HEALTH
EVALUATION AND DATA PROCESSING IN MEDICO-SOCIAL RESEARCH (FOR THE STUDENTS OF
GENERAL MEDICINE OF THE ENGLISH-SPEAKING MEDIUM). 2.
Marc De Hert, N. D. (2011). Prevalence and correlates of seclusion and restraint use in children and
adolescents: a systematic review. 10.