A variable becomes measurable at exactly one point: when you can say which questionnaire items produce its score, on what scale, and which statistical tool consumes that score. The artefact that records those decisions is a single wide table. Many Philippine departments require it in Chapter 3; the rest require the information scattered across three sections, which is harder. Either way, building the table first is the cheapest way to find out that a variable is not yet measurable.
The six columns
| Column | What goes in it |
|---|---|
| 1. Variable | The variable exactly as named in the statement of the problem |
| 2. Conceptual definition | One clause, with its citation |
| 3. Operational definition | What the score is, in this study |
| 4. Indicators | The dimensions you measure, each one a group of items |
| 5. Instrument items and scale | The item numbers, the number of points, and the interpretation ranges |
| 6. Statistical tool | The tool that answers the specific question this variable belongs to |
Columns 2 and 3 are the same pair that appears in the Chapter 1 definition of terms, and they must be worded identically in both places. What this table adds, and what the Chapter 1 list cannot carry, is columns 4 to 6: the bridge from a definition to an actual number.
Six filled-in matrices
1. Social media marketing practice, business administration
| Column | Entry |
|---|---|
| Variable | Social media marketing practice |
| Conceptual | An enterprise’s use of social media platforms to create and exchange offerings with customers [citation] |
| Operational | The customer respondent’s mean rating across Part II |
| Indicators | Content frequency; responsiveness; promotional offers |
| Items and scale | Part II, items 1 to 15, five items per indicator; four-point scale, 4 = always to 1 = never |
| Statistical tool | Weighted mean for the level; Pearson r against purchase intention |
2. Knowledge of standard precautions, nursing
| Column | Entry |
|---|---|
| Variable | Knowledge of standard precautions |
| Conceptual | Awareness of the minimum infection prevention practices applied in the care of all patients [citation] |
| Operational | The number of correct responses on the knowledge items, converted to a percentage |
| Indicators | Hand hygiene; personal protective equipment; sharps handling; waste segregation |
| Items and scale | Part II, items 1 to 20, dichotomous; scored 1 correct, 0 incorrect |
| Statistical tool | Frequency, percentage and mean score; Pearson r against reported practice |
Note that this variable is not on a Likert scale, and its row says so. Mixing a knowledge test and an attitude scale into one column is where many matrices quietly go wrong.
3. Extent of use of reading intervention strategies, education
| Column | Entry |
|---|---|
| Variable | Extent of use of reading intervention strategies |
| Conceptual | The planned instructional approaches a teacher applies to learners needing additional reading support [citation] |
| Operational | The teacher respondent’s mean rating across Part II |
| Indicators | Phonics instruction; guided oral reading; vocabulary building |
| Items and scale | Part II, items 1 to 18, six per indicator; four-point scale, 4 = always used to 1 = never used |
| Statistical tool | Weighted mean; Pearson r against pupil oral reading performance |
4. Oral reading performance, education
| Column | Entry |
|---|---|
| Variable | Oral reading performance |
| Conceptual | A learner’s accuracy and fluency in reading a passage aloud relative to grade-level expectations [citation] |
| Operational | The pupil’s level as recorded in the school’s existing reading inventory for the school year stated |
| Indicators | Not applicable; the level is taken as recorded |
| Items and scale | No researcher-administered instrument; the school’s recorded levels are used |
| Statistical tool | Frequency and percentage for distribution; Pearson r against extent of use |
A row built on existing records is legitimate and often stronger than a new instrument, but it needs the permission and the record’s date in the data gathering procedure.
5. Perceived service quality, hospitality
| Column | Entry |
|---|---|
| Variable | Perceived service quality |
| Conceptual | The guest’s judgement of overall service excellence, formed by comparing expectation with perceived performance [citation] |
| Operational | The guest respondent’s mean rating across the five dimensions in Part II |
| Indicators | Tangibles; reliability; responsiveness; assurance; empathy |
| Items and scale | Part II, items 1 to 25, five per dimension; five-point scale, 5 = strongly agree to 1 = strongly disagree |
| Statistical tool | Weighted mean per dimension and overall; Pearson r against satisfaction; ANOVA by frequency of visit |
6. Extent of adoption of internal control practices, accountancy
| Column | Entry |
|---|---|
| Variable | Extent of adoption of internal control practices |
| Conceptual | Processes providing reasonable assurance over reporting reliability, operational effectiveness and compliance [citation] |
| Operational | The owner or manager respondent’s mean rating across Part II |
| Indicators | Control environment; control activities; monitoring |
| Items and scale | Part II, items 1 to 21, seven per indicator; four-point scale, 4 = fully adopted to 1 = not adopted |
| Statistical tool | Weighted mean; Pearson r against perceived reliability; Kruskal-Wallis by years in operation |
Computing the interpretation ranges rather than copying them
Column 5 needs verbal interpretation ranges, and the commonest way students get them is by copying a table from a thesis with a different scale. Derive them instead. The class width is the range of the scale divided by the number of points.
Four-point scale. The range is 4 minus 1, which is 3. Divided by 4 points, the class width is 0.75. The ranges are therefore 1.00 to 1.75, 1.76 to 2.50, 2.51 to 3.25 and 3.26 to 4.00.
Five-point scale. The range is 5 minus 1, which is 4. Divided by 5 points, the class width is 0.80. The ranges are 1.00 to 1.80, 1.81 to 2.60, 2.61 to 3.40, 3.41 to 4.20 and 4.21 to 5.00.
Show that division in Chapter 3 in one line. It costs a sentence and it removes a question. Attach a verbal label to each band, and use the same labels in every table in Chapter 4, because a label that changes between tables is read as carelessness.
The four checks that catch a broken row
- Does every indicator have items? An indicator with no item numbers is a dimension you named but did not measure. Either write items for it or remove it from the conceptual framework as well.
- Do the item numbers add up? If column 5 says fifteen items across three indicators, the questionnaire must contain fifteen, and the parts must divide evenly unless you say why not.
- Does the statistical tool match the level of measurement? A mean on an ordinal single item is a shortcut many Philippine departments accept for composite scores but not for one item. Ask yours.
- Does every specific question have at least one row? Work from the statement of the problem downward. A question with no row in this table is a question with no method.
Check 3 is the one that most often changes a row rather than confirming it. Which tool a given question actually needs is set out in the piece on statistical treatment of data, and what to do when the distribution refuses to cooperate is in the piece on data that is not normally distributed.
Where the table goes, and what it feeds
Philippine manuals place it in Chapter 3, most often under the research instrument section or immediately after it, presented as a table with a number and a title. Some departments put it in an appendix instead and summarise it in the text. Either is acceptable; what is not acceptable is having the information nowhere.
| Column | Feeds |
|---|---|
| Indicators | The parts and subheadings of the questionnaire; the boxes of the research paradigm |
| Items and scale | The instrument section; the pilot test and the reliability computation |
| Interpretation ranges | Every table in Chapter 4 |
| Statistical tool | The statistical treatment section; the analysis you actually run |
The boxes of the diagram and the indicators in column 4 are the same list, which is why the research paradigm and this table have to be built together. Where the table sits among the other Chapter 3 sections is set out in the guide to writing Chapter 3 section by section, and how the resulting numbers are presented is covered in the guide to Chapter 4, table by table.
Build the table before you print the questionnaire
Almost every instrument that fails validation fails because a dimension had no items or an item belonged to no dimension, and both are visible in this table before a single copy is printed. Tesify builds the matrix from your variables and specific questions, computes the interpretation ranges for the scale you choose, and keeps the indicator names identical across the paradigm, the questionnaire and Chapter 4. The measurement decisions remain yours.
Build your operationalisation table free on Tesify
Frequently asked questions
What is operationalisation of variables?
It is the process of turning an abstract variable into something countable, by naming its indicators, the items that measure each indicator, the scale used and the statistical tool applied to the resulting score.
Is the operationalisation table the same as the definition of terms?
No. The definition of terms is a Chapter 1 list of defined words. The operationalisation table is a Chapter 3 matrix that adds indicators, item numbers, scales and statistical tools. The conceptual and operational definitions appear in both and must be worded identically.
Where does the table go in Chapter 3?
Most often within or immediately after the research instrument section, as a numbered table. Some departments place it in an appendix. Your research manual decides.
How many indicators should a variable have?
Two to five is normal. One indicator means the variable and the indicator are the same thing, which is fine but should be stated. More than five usually means two variables have been merged.
How do I compute the verbal interpretation ranges?
Divide the range of the scale by the number of points. On a four-point scale the range is 3 and the class width is 0.75; on a five-point scale the range is 4 and the class width is 0.80. Show the division in Chapter 3.
Can I copy an interpretation table from another thesis?
Only if it used the same number of scale points, and even then it is better to show your own computation. Copied ranges from a different scale are a common and easily spotted error.
Does every variable need indicators?
No. A variable taken from an existing record, such as a recorded performance level or an enrolment figure, has no indicators. Write “not applicable” in that cell and say where the record comes from.
What if I am using a published standardised scale?
The scale supplies columns 4 and 5 for you, and you cite it rather than inventing them. Do not renumber or reword its items, since that breaks the validation that made it worth using.
Do qualitative studies operationalise variables?
Not in this form. A qualitative study describes how a concept was identified in the data and which interview guide questions elicited it, which is the same logic without scales or statistical tools.
My adviser did not ask for this table. Should I still build it?
Build it for yourself even if it never appears in the manuscript. It is the fastest check that every specific question has a method and every indicator has items, and both are things a panel will test.