RESEARCH 2.6.

2.7.
Projective devices
Stoichiometric devices
2.2. Structured
2.3. Semi-structured
1. Focus observation
2. Concealment
INSTRUMENT Most Frequently Used Data Strengths
3. Duration

Instrument Collection Techniques 1. Good for measuring attitude Types of Observation

-gen. term for researches 1. Documentary analysis & most other content of 1. Structured
-use for measuring device (survey, 2. Interview Schedule interest -uses checklist (data collection
test, questionnaire, etc.) 3. Observation 2. Allow probing by the tool)
-to help distinguish between 4. Physiological Measurement interview -specifies behaviors of
instrumentation, consider that the 5. Psychological Test 3. Can provide in-depth info interest
instrument is the device & 6. Questionnaire 4. Allow good interpretative
instrumentation is the course of validity 2. Unstructured
action (the process of dev., testing 1. Documentary Analysis 5. Very quickly turnaround for -observes things as they happen
& using the device)  Use too analyze primary & telephone interview -conducts experiment w/out
-researchers (chooses which type secondary sources 6. Moderately high preconceived ideas about what
of instrument/s to use based on the  @ times, data are not measurement validity for will be observed
research question) available/ difficult to locate well-constructed & well-
2 Broad Categories  Available mostly: tested interview protocols. 4. Physiological Measures
1. Researcher-Completed (1)churches, (2)schools,  collection of physiological
Instrument (3)public/private offices, Weaknesses data from subjects
1.1. Rating Scales (4)hospitals/ community 1. In-person int. are expensive  more accurate & objective
1.2. Interview  Information gathered tend & time-consuming that any other data-
Schedule/guides to be incomplete & not 2. Perceived anonymity by collection methods
1.3. Tally Sheets definite respondents possible low
1.4. flow chart 3. Data analysis sometimes 5. Psychological Test:
1.5. Performance checklist 2. Interview Schedule time consuming for open- 1. personality inventories
1.6. time-&-motion logs  skill of interviewer ended items. -self reported measurements that
1.7. Observation forms determines if the assess the differences of
interviewee is able to 3. Observation personality traits/values of people
2. Subject-Completed express his thoughts clearly  observation guide/checklist -gathers info form a person thru
Instrument  conducted w/ single person (instrument used) questions/statements that requires
2.1. Questionnaires  Focus group interview  must be done quietly & responses/reactions
2.2. Self-checklist (conducted w/ gr. of inconspicuous manner so as -EX.: Minnesota Multiphasic
2.3. attitude scales people) to get realistic data Personality Inventory (MMP) & the
2.4. Personality inventories Edwards Personal Preference
2.5. Achievement/aptitude Types of Interview Schedule the ff. should be taken into Schedule (EPPs)
tests 2.1. Unstructured consideration:
 2. projective techniques variables being investigated -Q is a combination of 2/ more of standards addressed by an
-subject is presented w/ stimulus in the study types of Q assessment. For ex., if you
designed to be ambiguous/vague  descriptive info gathered administer an assessment that
in meaning.
-EX.: Rorschach Inkblot Test &
from diff. sources are called
indicators for the specific
VALIDITY & covers 3 standards, you
CANNOT infer from those results
Thematic Apperception Test variable & they are used in
making sure that the
RELIABILITY whether/ not a student is
proficient in 7th grade
6. Questionnaire (Q) content of the Q is valid mathematics (cuz that is NOT
 most common used  Valid indicator (supported Validity what you’ve assessed).
instrument in research by prev. studies done by  accuracy of inferences
 list of Q about a topic, w/ experts) drawn from an assessment Types of Construct Validity
spaces provided for the  degree to which the 1. Convergent Validity
response, & intended to be Types of Question assessment measures what it 2. Discriminant Validity
answered by a # of person 1. Yes/ No Type is intended to measure (na 3. Known-Group Validity
 less expensive -ex. do you have the right to refuse measure ba niya ang dapat 4. Factorial Validity
 yields more honest the call? yes/no? e measure?) 5. Hypothesis- Testing Validity
responses
 guarantees confidentiality 2. Recognition Type Types of Validity 1. Convergent Validity
 minimizes biases based on -alternative responses are provided 1. Construct validity  there is evidence that the
question-phasing mode & respondents simply choose 2. Face Validity same concept measured in
 researcher must make sure among the choices 3. Criterion-related Validity diff. ways yields similar results
Q is valid, reliable &  include 2 dif. test
unambiguous (critical point 3. Completion Type I. Construct Validity  diff. measures of same
in designing) -respondents are asked to fill in the  the assessment actually concept yield similar results
tables w/ necessary info measures what is designed to  researchers uses self-report
3 Types of Questionnaires measure (A is actually A) v.s observation (diff.
1. Closed-ended/ structured 4. Coding Type  refers to “Validity” measures)
2. Open-ended/Unstructured -Numbers are assigned to names,  what you’re trying to measure
3. mixture of close & open date & other pertinent data (i.e. IQ test is intended to
measure intelligence)
5. Subjective Type  ex.: you are trying to measure
Relationship of RRL to the -respondents are free to give student’s knowledge of the
Questionnaire opinions about issue of concern solar system. You don’t want to
 RRL& studies must have be measuring their ability to
sufficient info & data to read complex info text.
enable the researcher to 6. Combination Type  ex.: This is most commonly an
thoroughly understand the issue with regard to the breadth
2. Discriminant Validity statistical model called Types of Criterion-Related Validity  validity is about providing
 there is evidence that 1 factor analysis 1. Predictive evidence
concept is diff. form other 2. Concurrent  ECD boosts validity:
closely related concepts 1. What you want to know?
1. Predictive Validity (A predicts B) 2. How would you know?
 assessment predicts 3. What should the
performance on a future assessment look like?
assessment
Reliability
 how consistent a measure is
5. Hypothesis-testing validity
of a particular element over
 evidence that a research
a period of time, & between
3. Known- Group (KG) validity hypothesis about the
diff. participants
 group w/ established relationship between the
 is synonymous w/ the
attribute of the outcome of measures concept (variable) or
consistency of a test, survey,
construct is compared w/ a other concepts (variables),
observation, or measuring
group whom the attribute is derived from a theory, is 2. Concurrent validity (A correlates
device
not yet established supported w/ B)
 ex,: in a survey that used  assessment correlates w/
4 General Estimators
Questionnaire (Q) to explore II. Face Validity other assessments that
1. Inter-Rated/ Observer
depression among 2 gr. of  when an individual (& or measure the same
Reliability
patients w/ clinical diagnosis researcher) who is an expert on constructs
2. Test-Retest Reliability
of depression & those w/out. the research subject reviewing
3. Parallel-Forms Reliability
it is expected in KG that the the questionnaire (instrument)
4. Internal Consistency
construct of depression in concludes that it measures the Valid Inferences
Reliability
the Q will be scored higher characteristic/trait of interest  “Validity” (closely tied to the
among the patients w/ purpose/ use of an
Inter-Rated/ Observer Reliability
clinically diagnosed III. Criterion-Related Validity assessment)
 used to assess the degree to
depression than those w/out  Assessed if interested i
which diff. judges/raters
the diagnosis. determining the relationship of DON’T ASK: “Is this assessment
agree in their assessment
scores on a test to a specific valid?”
decisions
4. Factorial Validity criterion. ASK: “Are the inferences I’m
 useful cuz human observers
 empirical extension of  a measure of how well a making based on this assessment
will not necessarily interpret
content validity questionnaire findings stack up valid for my purpose?”
answers the same way
 it validates the contents of against another
 raters may disagree as to
the construct employing the instrument/predictor Evidence-Centered Design (ECD)
how well certain responses/
material demonstrate
 knowledge of the responses are. In general,  ex.: a personality test may individuals, the total score
construct/skill being correlation coefficient (r) seem to have 2/more for each “set ” is computed,
assessed. values are censured good if questions that are asking the & finally the Split-half
 uses2 individuals to mark/ r > 0.70 same thing. If the reliability is obtained by
rate scores of psychometric participants answers these determining the correlation
test, if their scores/ratings Parallel Form of Reliability similarly, then the internal between the 2 total “set”
are comparable then inter-  obtained by administering consistency reliability is scores
rated reliability is confirmed diff. versions of assessment assumed correct.
tool (both version must 2 Types of Internal Consistency
Test-Retest Reliability contain items that probe the Reliability
 Obtained by administering same construct, skill, 1. Average inter-item
the same test twice over a knowledge, base, etc.) to correlation
period of time to a group of the same indi 2. Split-half reliability
individuals.  the scores from the 2
 the scores form time 1 & 2 versions can be correlated Average inter-item correlation
can be correlated in order in order to evaluate the  obtained by taking all the
to evaluate the test for the consistency of results across items on the test that probe
stability over time alternate versions the same construct (e.g.
 this is the final sub-type & is  To evaluate the reliability of reading comprehension)
achieved by giving the critical thinking assessment:  determining the correlation
same test out at 2 diff. times you might create a large set coefficient for each pair of
& gaining the same result of items that all pertain to items, & finally taking the
each time critical thinking & then slowly average of all these
 to assess student learning in split the questions up into 2 correlation coefficients
psychology could be given sets, which could represent  this final step yields the
to a group of students twice, the parallel forms. average inter-item
w/ the second correlation
administration perhaps Internal Consistency Reliability
coming a week after the 1st.  used to evaluate the Split-half reliability
The obtained correlation degree to which diff. test  process: “splitting in half” all
coefficient would indicate items that probe the same the items of a test that are
the stability of the scores. construct produce similar intended to probe the same
 it is measures by having the results area of knowledge (e.g.
same respondents  this looks at the items w/in WW II)in order to form 2
complete a survey at 2 the test, to assess the “sets” of items
different points in time to internal reliability between  the entire test is
see how stable the items. administered to a group of
 POPULATION,  extent to which the
measurement tool provides
 what happens if the
instruments use change
2. Degree of Precision desired
by the Researcher
PARAMETER & consistent outcomes if the
measurement is repeatedly
during the course of a study
 if different groups of
 a larger sample size will result
in greater
SAMPLE perform participants interpret your precision/accuracy of results
questions differently? 3. Types of Sampling
Population
Regression towards the mean Procedure
 composed of
 if you start w/ extreme  probability sampling (by
persons/objects that possess
groups, they are likely to be chance) utilizes smaller sizes
some common charact
less extreme at another time than NON-probability
 eristics that are of the same
point sampling (by choice)
interest of the researcher
Selection 4. The use of Formula
 are participants in your study a. Slovin’s Formula
Parameter
Threats to Internal Validity somehow diff. than those 𝑁
 numeric/other measurable 𝑛=
1. history who are not? 1 + 𝑁^2
characteristics forming 1 of
2. maturation Mortality
a set that defines a system
3. testing  those who drop out of your b.Calmorin’s Formula
 sets the condition of its
4. instrumentation study could be qualitatively
operation 𝑁𝑉 + [(𝑆𝑒)2 𝑥(1 − 𝑝)]
5. regression towards mean different than those who 𝑆𝑠 =
Sample 𝑁𝑆𝑒 + [𝑉 2 𝑥 𝑝(1 − 𝑝)]
6. selection remain
 subset of the entire
population or a group of 7. mortality
PARTICIPANTS OF Sampling
individuals
 represents the population History THE STUDY  a way of getting a
representative portion of the
 serves as the respondents of  there is something about the
experience & history of 1 target population
the study Ways to Determine Sample Size
group that could lead to  needed regardless of the
Methodology
differences? research design for practical
 is the systematic, theoretical Factors to consider in determining
Maturation reasons especially when the
analysis of the methods Sample sizes:
 natural development population is large
applied to a field of study 1. Homogeneity of the
influence the association?  at TOTAL population of 100/
Validity population
Testing less may not need to be
 the degree to which  the higher the degree of
 does actual participation in sampled
research instrument gauges variation w/in the
the study change the  greater than 100 needs
 what it is supposed to population, the smaller the
behavior/ responses? sampling for desirable
measure? sample size that can be
Instrumentation effectiveness, efficiency &
Reliability utilized economy in data gathering
 area, otherwise the results Simple Random Sampling  ex.: the population of
Research Design obtained will be erroneous.  all members of the students in 1 school is listed
1. Descriptive 4. Characteristics to be population are given a alphabetically & numbered
2. Experimental observed may occur rarely chance to be selected consecutively. From the list,
in the population (ex.  selection is done by draw lot the sample to be taken is
2 Strengths (Sampling) teacher over 30 yrs. of / the use of the table of the name that falls every nth
teaching experience) random numbers. in the list until the desired
1. time, money & effort are 5. Complicated sampling number of sample is
minimized plans are laborious to Get a list of the total population completed
 number of respondents, prepare 1. cut pieces of paper to small
subjects/ items to be studied sizes (1x1 in)that can be Sampling Scheme
becomes small yet they 2 General Types of Sampling rolled  used when there is a ready
represent the population 2. put a number on each list of the total population
 data collection, analysis & 1. Probability sampling / piece of paper  the steps:
interpretation are lessened Scientific sampling corresponding to each o get the list of the total
 gives every member of the number of the total population
2. Sampling gives more population equal chance to population o divide the total
comprehensive information be selected as part of the 3. Roll each paper of the population by the
 small sample representing a study. number and put them in the desired sample size to
big population, a thorough a box get the sampling
study yields results that give 1. Non Probability Sampling/ 4. Shake well the box to give interval
more comprehensive info Non- Scientific sampling equal chance for every
that allows generalization &  opposite of probability number to be chosen a Stratified Random Sampling
conclusion  researcher’s judgment sample  dividing the population into
determines the choice of 5. pick out 1 rolled paper at a strata & drawing the sample
5 Weaknesses (Sampling) sample time & unfold it at random from each
1. Due to limited number of 6. Record the number of the division
data source, detailed sub 5 Types of Probability Sampling/ unrolled paper
classification must be Scientific Sampling 7. Repeat picking out until the Used when there is a ready list of
prepared w/ utmost care 1. Simple Random Sampling desired number of sample is the population whose members
2. incorrect sampling design/ 2. Systematic Sampling completed. are categorized as students,
incorrectly following the 3. Stratified Random Sampling Systematic Random Sampling farmers, and fisherman. The steps
sampling plan will obtain 4. Cluster Sampling  chooses nth name of a to follow are:
results that are misleading 5. Multi-stage Sampling population as the sample 1. get the list of the total
3. Sampling requires an expert  it entails using a list of the population
to conduct the study in an population & deciding how 2. decide on the sampling
the nth name is chosen size/ the actual percentage
 of the population to area that will serve as the Example: characteristics/ prescribed
considered as sample. cluster 1st level. 3 municipalities/province criteria of the research
3. get an equal proportion of a 2. select the sample wi/in the 2nd level. 2 district/ municipality
sample from each group.
Example:
area or cluster by
systematic/ stratified
3rd level. 4 barangays/ districts
4th level. 100
FINDING THE
500 pupils x .20 = 100
200 teachers x .20 = 40
sampling depending on the
availability of the info
respondents/barangay
SAMPLE SIZE
150 parents x .20 = 30 *use any of the sampling tech. Slovin’s Formula
Total Sample= 110 Multistage Sampling given earlier in the arriving at the  use to calculate the sample
*get the 110 respondents by simple  done by stages: 2,3,4 as the desired sample number. size(n) & the margin of
random sampling/ systematic case may be depending on error(e)
sampling the number of stages 3 Types of Non Probability  random sampling tech.
sampling is made Sampling/ Non- Scientific sampling formula to estimate
Cluster Sampling  Population is grouped by 1. Purposive Sampling sampling size
 uses a group as sample hierarchy from which 2. Incident/ Accidental
rather than an indi sampling is done in each Sampling 𝑁
𝑛=
 ex.: the population may be stage. 3. Quota Sampling 1 + 𝑁^2
the parents in 1 school  ex.: the population to be
district. the parents may be studied consists of the Purposive Sampling Example:
grouped by barangay w/in personnel in the public  samples are choses based the parameter of the
the district or by those in the elementary schools in the on the judgment of the population is 8K at 2 % margin
east, west, north, & south of country. Samples have to researcher who determines of error or 98% accuracy
the district. be take form the national, an individual, as sample for 𝑁
 differs from stratified regional, provincial, district, possessing special 𝑛=
1 + 𝑁^2
sampling that includes all & school levels. characteristics of some sort
the strata in the sampling Steps followed in this scheme: 8,000
=
process 1. decide on the level of Incidental/ Accidental Sampling 1 + 8,000 (0.02)2
analysis to be studied such  this design is used to take
Used when the population od national, regional, samples who may be 8,000
=
HOMOGENEOUS but scattered provincial, district, available/ the nearest at 1 + 8,000 (0.0004)
geographically in all parts of the barangay, etc. the time of data gathering 8,000
=
country & that there is no need to 2. select the sample at the 1 + 3.2
include all in the sampling. Steps in next level. ex.: if the survey is Quota Sampling 8,000
=
the SCHEME: provincial, the survey should  popular for opinion research 4.2
=1,905
1. decide on the sample size. start at the municipal level,  made by looking for indi
select the geographical then district, then barangay, that possess the required
etc.
Calmorin’s Formula
 Population of more than
100, sampling is a must
 follows formula for scientific
sampling, illustrations &
examples (Calmorin &
Calmorin, 1995)
𝑁𝑉 + [(𝑆𝑒)2 𝑥(1 − 𝑝)]
𝑆𝑠 =
𝑁𝑆𝑒 + [𝑉 2 𝑥 𝑝(1 − 𝑝)]

When Ss=Sample Size

N= total no. of population
V= standard values (2.58) of 1 %
level of the probability with 0.99%
reliability
Se=Sampling error
P= largest possible proportion (0.50)

Example:
The total population is 500 has a
standard value of 2.58 at 1% level
of probability & 99% reliability. the
sampling error is 1% (0.01) & the
proportion of a target population is
50% (0.50)
given:
N= 500
V=2.58
Se=0.01
P=0.05
𝑁𝑉 + [(𝑆𝑒)2 𝑥(1 − 𝑝)]
𝑆𝑠 =
𝑁𝑆𝑒 + [𝑉 2 𝑥 𝑝(1 − 𝑝)]
500(2.58) + [(0.01)2 𝑥(1 − 0.5)]
𝑆𝑠 =
500(0.01) + [2.582 𝑥 0.5(1 − 0.5)]
1290.00005
𝑆𝑠 =
6.6641