Functions/Uses of Statistics or one that might have happened by chance

in a study
- Helps in providing a better understanding
and exact description of a phenomenon of Differences between Descriptive & Inferential
nature. Statistics:
- Helps in proper and efficient planning of a
- With descriptive statistics, one is simply
statistical inquiry in any field of study.
describing what is or what the data shows
- Helps in collecting and classifying an
while with inferential statistics one tries to
appropriate set of quantitative data.
reach conclusions that extend beyond the
- Helps in presenting complex data in a
immediate data
suitable tabular, diagrammatic and
- Descriptive statistics uses graphical
graphic form for an easy and clear
numerical descriptions to give a picture of a
comprehension of data.
data set while inferential statistics uses
- Helps in understanding the nature and
mathematical probabilities to make
pattern of variability of a phenomenon
generalizations about a large group based on
through quantitative observations.
data collected from a small sample of that
- Helps in drawing valid inference, along
group
with a measure of their reliability about the
population parameters from the sample
o Population – portion of the universe that
data.
can be reached by a researcher
- Helps in understanding statistical
- includes all individuals with certain
techniques for the purpose of making
specified characteristics in a target
informed decisions that affect our lives and
community/locale or setting
well-being.
- the group to which the researcher would
like the results of a study to be generalizable
o Statistics – branch of mathematical science o Sample – the group or portion of a
that deals with the scientific collection, population on which information is
classification, organization, description, obtained, preferably selected in a such a
analysis & interpretation of data obtained way that it represents the population
from surveys and experiments o Data –facts, observations, and information
- deals with prediction and forecasting based that come from investigations
on data
s
Types of Data:
Two Types of Statistics
1. Measurement data/ quantitative data – the
result of using some instrument to measure
1. Descriptive Statistics - to describe the basic something (test score, weight, length, height,
features of the data in a study speed)
- to present quantitative descriptions in a 2. Categorical data/ nominal or qualitative
manageable form data – provide categories for sorting or
- provides summaries about the sample & classifying objects or events on the basis of
measures and form the basis of virtually some quality like major field (Sociology,
every quantitative analysis of data Psychology, Colors, Mathematics, Sex)
- includes the Measures of Central Tendency 3. Ordinal Data - provide a system for
& Measures of Dispersion ranking observations from most to least or
least to most
- includes the position of individuals
2. Inferential Statistics – to make inferences
finishing a race (first, second, third); social
or generalizations on a population based
class position (upper, middle, lower) and any
upon a sample and used to test hypothesis
data involving scales that organize
and evaluate estimates
observations in terms of categories (ex. Very
- to make judgments or conclusions on the
favorable, favorable, neutral, unfavorable,
probability that an observed difference
very unfavorable)
between/among groups is a dependable one
- The rule for assigning numerals to ordinal - If these cannot be controlled, these must be
data categories is based on ordering taken into consideration when interpreting
observations in a descending or ascending results.
order. 7. Intervening – processes that are not directly
4. Interval data - provide categories which observable which like extraneous variables
consist of equal intervals which indicate that can alter the results of research.
the distance of each interval is known. - In Language teaching for example, these are
Example of categories which show interval usually inside the subjects’ heads including
data are: ruler (the distance between 1 inch and various language learning processes which
2 inches is exactly the same as the distance the researcher cannot observe
between 7 and 8 inches), age, number of years - If teaching technique is the independent
of formal education, weekly earnings, etc. variable and mastery of objectives is the
dependent variable, then the language
o Variable – property of an object or event learning processes used by the subjects are
that can take on different values. Example, the intervening variables
college major is a variable that takes on values - include motivation, fatigue, boredom, and
like mathematics, computer science, English, any other factor that arises during the
psychology course of the research either on the part of
the researcher or the respondents
Types of Variable
8. Moderator – affect the relationship
between the independent and dependent
1. Continuous – consists of an infinite
variables by modifying the effect of the
continuum of points and provides exact
intervening variable/s.
measures of the amount of a characteristic
- Unlike extraneous variables, moderator
present, in theory, any value between the
variables are measured and taken into
lowest and highest points on the
consideration
measurement scale
- Typical moderator variables in language
2. Discrete – provides counts of the number of
acquisition research, when these are not the
observations appearing in a finite set of
focus of the study, include, age, gender,
categories (gender, (male/female), college
culture or language proficiency of the
class (freshmen/sophomore/junior/senior)
subjects
3. Independent/ experimental or predictor
variable - is manipulated, measured or
selected by the researcher as an antecedent Measures of Location/Center /Central
condition to an observed behavior. Tendency or Averages
- In a hypothesized cause-and-effect
relationship, the independent v is the cause 1. Median - called the counting median
& dependent v is the outcome or effect. - most stable measure of central tendency
4. Dependent/ criterion variable - that is not - middlemost score of the distribution
under the experimenter’s control – the data of scores arranged from highest to
- Variable that is observed and measured in lowest when N is odd, the middle
response to the independent variable. most score is the counting median
5. Control – Variables which are not - when N is even, the counting median is the
measured in a particular study must be held average of the two middlemost scores
constant, neutralized/balanced or
N = even
eliminated so that these will not have a

( )
TH
biased effect on the other variables N
x=
6. Extraneous - factors in the research 2
environment which may have an effect on
the dependent variable/s but which are not N = Odd
controlled
( )
TH
- may damage a study’s validity making it N +1
X=
impossible to know whether the effects were 2
caused by the independent variable
2. Mean - most reliable average 3. Area Sampling - One special type of
- called arithmetic average cluster sampling, where pieces of
- the average of the given scores using the geographical areas such as districts,
formula: housing blocks or townships are selected
- could be one-stage, two-stage, or multi-stage
where: Σx is the sum of scores
- Generally used by Gov’t agencies and
N = the number of cases agricultural statistics.
M = is the mean 4. Convenience sampling - process of
including whoever happens to be available
x = Σx at the time called "accidental" or
N "haphazard" sampling.
Weighted Mean:
Wm = Σx W Other Non-probability Sampling Methods
ΣW 1. Intensity sampling: selecting participants
]
who permit study of different levels of the
3. Mode - used when one wants to see the research topic
trend of the scores 2. Homogeneous sampling: selecting
- called rough mode participants who are very similar in
- the score with the greatest frequency in the experience, perspective, or outlook
distribution 3. Criterion sampling: selecting all cases that
meet some pre-defined characteristic
SAMPLING TECHNIQUES 4. Snowball sampling relies upon respondent
referrals of others with like characteristics
o SAMPLING - Procedure by which some
members of a given population are selected
as representatives of the entire population Factors to Consider in Sample Design
- representatives selected for a study whose a) Research objectives
characteristics exemplify the larger group b) Resources
from which they were selected c) Knowledge of target population
o POPULATION - larger group from which d) Degree of accuracy
individuals are selected to participate in a e) Time frame
study f) Research scope
g) Statistical analysis needs

1. Simple random sampling - a method of

probability sampling in which every unit has The Sampling Process
an equal non zero chance of being selected a) Define the Population
for the sample. b) Develop Sampling Frame
c) Select a Sampling Method
Methods of selecting random sample d) Determine the Sample Size
e) Execute the Sampling Process
a. Lottery Method
b. Tables of Random Numbers Systematic Random Sampling or Quasi-
Random Sampling
- a method of probability
2. Stratified random sampling - method of
in which the defined target population is
probability sampling in which the population
ordered and the 1 unit of sample is selected
is divided into different subgroups and
at random and rest of the sample is selected
samples are selected from each of them.
according to position using a skip interval
Steps: (every Kth item)
 All units of population are divided into K = N/ n
different stratas in accordance with their Where, K = Sampling/ Skip interval
characteristics N = Universe/ Population Size
 Using random sampling, sample items are n = Sample Size
selected from each stratum.
 MULTISTAGE RANDOM SAMPLING

- Used in large scale investigations

- First stage - preparation of large sized
sampling units
- Randomly selecting a certain number
- Second stage -Another list prepared from
them
- Sub-samples drawn by random sampling
Purposive sampling
- the process whereby the researcher selects a
sample based on experience or knowledge
of the group to be sampled called
"judgment" sampling

Quota sampling
- the process whereby a researcher gathers
data from individuals possessing identified
characteristics and quotas