Accuracy Refers To Obtaining The Closest Prediction To The Truest Statistic Given The Hypothesis And Data Presented

Unlock access to all the studying documents.

View Full Document

M.A, Hassim, 201133806, BLB17X7

Accuracy refers to obtaining the closest prediction to the truest statistic given the

hypothesis and data presented. There are many statistical measures taken to

improve accuracy of results in a given regression depending on the formula and

nature of the data as well as the relationship between the data. Beins & McCarthy

(2012) refers to an aspect of accuracy being related to the sample size of the data as

opposed to the size of the population. He goes on to say that accuracy is

proportional to the square root of the sample size and is sensitive to changes in the

characteristics of the data.

Greener (2008) refers to action research is social research which is conducted by a

team of “action researchers” who participate in solving social problems, such

problems can be found in the fields of politics, education, environmental safety etc.

ANOVA is the abbreviation for analysis of variance. The following definition of

analysis of variance is provided by Scheffe (1959:3): “The analysis of variance is a

statistical technique for analysing measurements depending on several kinds of

effects operating simultaneously, to decide measurements or observations may be in

an experimental science like genetics or a nonexperimental one like astronomy. A

theory of analysing measurements naturally has implications about how the

experiment should be planned or the observations should be taken. “

A baseline point or model which is used as a foundation to compare other models or

to refer to as a reference point. Greener (2008) refers to a baseline model in the

subject of multi-level modelling, where there are different predictors for each level

such as student characteristics and classroom characteristics which can be

examined also called an unconditional model. The baseline model is can be used to

show the effect of a variable on the model and the effect of a variable when it is

excluded from the model.

Keller (2006:34) defined the bell curve as follows: “The bell curve and the normal

curve are two names for the same thing. Originally, it was called the normal curve of

errors. This term comes from some of the earliest work in statistics, which focused

on predicting a son’s adult height from the height of his father. The errors from these

predictions formed a bell-shaped curve. Other characteristics were then predicted.

The errors from those predictions also formed bell-shaped curves. Remarkably,

M.A, Hassim, 201133806, BLB17X7

many common characteristics did the same thing. That bell-shaped curve was then

named the normal curve of errors.”

Benchmarking refers to the point of comparison between a unit and its

counter parts. It is a method of pinpointing where a unit is on a

scale given the place of other units. Ross (2007) refers to

benchmarking in a study done on Reflective Self-Positioning (RSP).

This study documents four aspects of members behaviour to help each

participant know where they stand within the program and allows them

to benchmark themselves against other participants. For example,

health regulations require restaurants to meet certain criteria to be

benchmarked before they are allowed to continue running.

A bias in terms of research is usually when an error occurs or when the actual results

deviate from the expected results, this may occur due to problems in the sample

selection process as well as other reasons (Henry, 1990). There are different types

of bias; a sampling bias is one of them. A sampling bias in research occurs when the

calculated statistic isn’t a true reflection of the population statistic. This could render

a flaw in the experiment; steps need to taken to remedy the error.

Case study research involves answering a research question involving a particular

population outside laboratory practices. This type of research is usually descriptive

where a particular cause to the research question is described or the end result of a

particular effect on a population is described (Greener, 2008).

Causal relationship explains the relationship of the effect or impact a particular cause

has on a country or community or the particular subject in question. In Hao-Yen

Yang’s study, the effect of GDP on the energy consumption of Taiwan was

investigated and this causal relation was explained in terms of what does an

increase or decrease in GDP do the energy consumption of the country (Greener,

2008).

Causality can be described simply as the effect caused by one variable upon another

i.e. the effect of the independent variable on the dependent variable. Greener (2008)

suggests that many assumptions are made regarding causality but in fact there is no

strong motivation that the independent variable actually did have a causal effect on

the dependent variable. For example one might say that performance bonuses make

M.A, Hassim, 201133806, BLB17X7

employees to work harder but there are other factors that are not taken into account

that may have a stronger effect on employees working harder.

The measures of central tendency are namely: the mean, median and mode. The

mode. The mean is the average of all the values, the median is the midpoint of all

the values and the mode is the most frequently occurring value. According to

Greener (2008) the measure of central tendency is a single figure so it is not

represented on a graph or chart. The measures of central tendency are usually

displayed in the form of a box plot where the interquartile range can be shown as

well.

The chi-square analysis, referred to as the chi-square test which allows the testing of

the deviations of actual frequencies from expected frequencies. Greener (2008)

describes where the chi-square test should be used; the chi-square test is used to

test the statistical significance which is known as the probability level or p. The

expected value is calculated of each cell in the contingency table and works out the

differences between the given values and the expected values and sums up the

differences.

Hill and Lewicki (2007:16) proposed the following explanation of cluster analysis:

”cluster analysis is an exploratory data analysis tool which aims at sorting different

objects into groups in a way that the degree of association between two objects is

maximal if they belong to the same group and minimal otherwise. Given the above,

cluster analysis can be used to discover structures in data without providing an

explanation/interpretation. In other words, cluster analysis simply discovers

structures in data without explaining why they exist.” Clustering is used on a day to

day basis simply by grouping and categorising items and people etc.

Cohort analysis is an analysis that is done on a population which has been divided

according to common behavioral traits or experiences shared within ones lifespan.

Such a study in terms of business is essential as it gives the company of interest the

opportunity to understand the consumers’ needs as each consumer is grouped

according to a particular characteristic and thus observe trends found within that

particular population (Beins & McCarthy, 2012).

M.A, Hassim, 201133806, BLB17X7

Conceptual framework is a definition of your study where each term, concept, theory,

variable of the study is defined. This is the main structure of the study and what the

researcher plans to do and what the expected outcomes may be (Greener, 2008).

Confidence interval is a statistical parameter representing the statistical uncertainty

of a population sample or set of data. An example would be the 95% confidence

interval implying a 5% uncertainty in sample data (Beins & McCarthy, 2012).

Confidence level is a percentage indication of all the possible samples to be

indicated in the true population. This gives an indication of the surety or certainty of

the confidence interval data. A confidence level of 95% shows the researcher is 95%

certain that the sample population data has a 5% confidence interval (Greener,

2008).

Confidence limits Is a range of specified numbers giving an indication of the

parametric mean. This shows a particular parameter lies within a specific range

definitely. This is the upper and lower values of the range (Greener, 2008).

Confirmability is to assess the truth, reliability of the data.in qualitative research,

confirmability is to actually assess the true meaning of data obtained. Meaning of the

respective data is assessed through a coding or meaning making system. The

coding system implies a particular code has a certain meaning. the meaning making

system employs looking for the meaning thorough all ones attention (Greener, 2008).

Construct or constructive research is research based on theories, hypothesis and

case studies. This type of research enables the researcher to test theories. It

provides solutions to practical and theoretical problems. This type of research solves

problems through the design of diagrams, organizations, plans and models (Greener,

2008).

Construct validity is an indication of how valid or reliable are the data obtained from a

particular test being used based on the properties of the test. Is the test actually

measuring what is supposed to measure? If the theory of particular test says it

should measure the amount of light passing through a prism, does the theory behind

the test actually happen and is the test actually measuring the amount light passing

through a prism (Greener, 2008).

M.A, Hassim, 201133806, BLB17X7

Constructionism is the understanding of a particular field by being exposed to the

multiple problems it has. This problem solving method allows the researcher to gain

insight into a particular field by understanding all the problems it entails. This then

gives the researcher the opportunity to design its own solution to the problem at

mind (Greener, 2008).

Constructivism in essence is the perceptions and views people develop and

construct an understanding of the world as they view it from the experiences and the

reflection of those experiences in their life-time. Greener (2008:18) points out the

how a constructivist views a team: “A constructivist view of a team is that every time

team members interact, they have a concept of team which is there in their minds

and which can alter over time depending on how they interact, but does not have an

independent reality.”

Content analysis can be either quantitative or qualitative in nature though it is more

often quantitative than qualitative. Content analysis is basically the description and

analysis of written or spoken content as well as images such as; books, interviews

and visual imagery from which relationships and patterns can be determined and

interpreted. Holsti (1969:14) offers a wider definition of content analysis as, “any

technique for making inferences by objectively and systematically identifying

specified characteristics of messages”. There are many other definitions by authors

that specifically relate to their respected fields.

Content validity refers to a measure or assessment of the subject matter and the

how well tests can be performed with this subject matter. The subject matter should

be able to be applied and put through a variety of tests within its domain. Content

validity is important as it helps establish whether the samples can be interpreted

(Litwin, 2003).

A continuous variable refers to a variable that falls anywhere between two points

regardless of how small the difference is between those two points. It can be used to

measure statistics like height and weight; referring to a specific measurement that

falls in a particular category. For example – if A is taller than B but C is both taller

than B and shorter than A; then C is a continuous variable (Beins & McCarthy, 2012).

A Control group is a group within an experiment or study. In the experiment there will

be an experimental groups or groups that the researchers will manipulate and

M.A, Hassim, 201133806, BLB17X7

experiment with variables within those groups. Greener (2008) refers to a control

group as a secondary group of similar subjects used for comparison. The control is

the group that is untouched with the variables left un-tempered and is used as a

benchmark in order to compare the experimental groups to. Control groups are

important in an experiment as they help guide the experiment.

A correlation is an idea that tests a relationship between variables rather than cause

and effect between the changes in the variables. Correlation looks to identify a

relationship through simple association and the strength of the relationship. For

example, does anxiety of public speaking have anything to do with the amount of

experience spent doing public speaking? The relationship between experience and

public speaking is determined through strength of correlation (Beins & McCarthy,

2012).

Critical realism is the perception or theory of studying natural and human social

sciences. The assessment of your data and determining whether it is realistic and

pragmatic in reality. This serves to establish the difference between what is scientific

belief and actual reality. This approach is widely used in philosophical approaches to

research (Johnson & Duberley, 2000).

Cronbach’s Alpha is a statistical measurement of the internal reliability and

consistency. It is also known as the Cronbach alpha reliability coefficient test.This

test is commonly used in psychology field. Cronbach’s alpha also measures the

correlation or how close a relationship there is between certain items in a group

(McKnight & Elias, 2003).

Data refers to a collection of information used to carry out an analysis. It can be

quantitative or qualitative and provides information on a posed hypothesis or

investigation. Data can be both numeric and non-numeric. It undergoes

transformation to allow the investigation to open up and shed light on the answers to

questions of research. This is data- processing (Greener, 2008).

The definition of data mining according to Hill and Lewicki (2007:23) is as follows:

“Data Mining is an analytic process designed to explore data (usually large amounts

of data – typically business or market related – also known as “big data”) in search of

consistent patterns and/or systematic relationships between variables, and then to

validate the findings by applying the detected patterns to new subsets of data. The

M.A, Hassim, 201133806, BLB17X7

ultimate goal of data mining is prediction – and predictive data mining is the most

common type of data mining and one that has the most direct business applications.

The process of data mining consists of three stages: (1) the initial exploration, (2)

model building or pattern identification with validation/verification, and (3)

deployment (i.e., the application of the model to new data in order to generate

predictions).”

The deductive approach or method of research starts from taking more broad

observations which narrows in to become more specific. According to Greener

(2008) regarding the deductive approach; the approach begins by taking a general

overview of the theory of the topic being researched as well as looking at literature

that relates to that theory and thereafter to test the theory. For example one might

want to research the effect of school students going to parties on their results and

then further narrow that down to the effect of grade 12 students from a certain high-

school on their results.

Degrees of freedom is a statistical measure of the estimation of the final results in a

calculation and how they may vary within a set parameter and constraints that have