5 Validated Scales You Can Field for Free This Week

Building a survey from scratch to measure a complex psychological or usability trait is a fast track to unreliable data.

Instead of debating whether your question wording introduces bias, you can borrow tools that have survived decades of academic scrutiny.

Validated scales give you a proven mathematical model for translating subjective human feelings into hard, defensible numbers.

Many of the most rigorous instruments in behavioral science and usability testing are open access and cost nothing to deploy.

Here are five free validated survey scales you can drop into your next research project.

Measure system usability using SUS

The System Usability Scale (SUS) has been a standard in usability testing since John Brooke published it in 1986. Originally created as a quick measure for electronic office systems, it has survived the transition to the modern web because it measures the user's subjective perception of a system, regardless of the underlying technology. It consists of 10 statements evaluated on a five-point Likert scale, ranging from Strongly Disagree (1) to Strongly Agree (5).

A common mistake is treating the raw SUS score as a direct percentage. The scoring system is intentionally complex to account for acquiescence bias - the human tendency to agree with statements rather than disagree. To counteract this, the SUS alternates between positive statements (like I think that I would like to use this system frequently) and negative statements (like I found the system unnecessarily complex).

Because of this alternating structure, you cannot simply add up the numbers the user selects. You must normalize the responses so that every item contributes to the final score on a scale of 0 to 4, where 4 is always the most positive response.

Follow these numbered steps to calculate the SUS score from 0 to 100 using 10 Likert items:

1. Convert odd-numbered items: For questions 1, 3, 5, 7, and 9 (the positive statements), take the user's raw response and subtract 1. If a user selects 5 (Strongly Agree), the converted score is 4.
2. Convert even-numbered items: For questions 2, 4, 6, 8, and 10 (the negative statements), subtract the user's raw response from 5. If a user selects 5 (Strongly Agree with a negative statement), the converted score is 0.
3. Sum the converted scores: Add the normalized values from all 10 items together. This will give you a total raw score ranging from 0 to 40.
4. Calculate the final metric: Multiply that total sum by 2.5. This scales the result to a maximum possible score of 100.

Once you have the final score, interpretation requires context. A score of 100 is perfect, but a score of 50 is not average. Decades of historical data show that the average SUS score across all software is roughly 68. If your application scores a 68, it sits at the 50th percentile for usability. Scores below 50 indicate severe usability issues that require immediate attention.

When setting up the SUS in your survey tool, ensure all 10 items appear on a single page or screen. Breaking the scale across multiple pages increases cognitive load and interrupts the rhythm of the alternating positive and negative statements. If a respondent skips an item by mistake, standard practice is to assign that missing item a neutral raw score of 3 before running your calculations.

PHQ-9: screening > clinical diagnosis

The Patient Health Questionnaire-9 (PHQ-9) is the standard instrument for measuring the severity of depression in both clinical and research settings. Developed by Drs. Robert L. Spitzer, Janet B.W. Williams, and Kurt Kroenke, it directly maps to the diagnostic criteria for major depressive disorder. Pfizer holds the copyright, but the scale is freely available for use, reproduction, and distribution.

The PHQ-9 asks respondents how often they have been bothered by nine specific problems over the last two weeks. The response options use a four-point frequency scale: Not at all (0), Several days (1), More than half the days (2), and Nearly every day (3). The total score ranges from 0 to 27.

Because it measures sensitive medical information, researchers must understand the boundary between data collection and medical intervention. The PHQ-9 is a screening tool, not a standalone diagnostic instrument. A high score indicates a high severity of depressive symptoms, but a clinical diagnosis requires a structured interview by a licensed healthcare professional to rule out other causes, such as physical illness or bereavement.

The true value of the PHQ-9 lies in its established severity brackets. Rather than analyzing the raw numerical score in isolation, researchers categorize respondents into clinical severity groups based on strict cutoffs.

Here are three full examples with wording of PHQ-9 severity categories and clinical scoring cutoffs:

• Mild Depression (Score 5 to 9): A respondent scoring a 7 might report feeling tired Nearly every day (3 points), having trouble sleeping More than half the days (2 points), and experiencing poor appetite Several days (2 points). In a clinical setting, this range typically prompts watchful waiting and a follow-up, rather than immediate active treatment.
• Moderate Depression (Score 10 to 14): A respondent scoring a 12 has crossed the threshold where clinical intervention is usually considered. They might report feeling down or hopeless More than half the days (2 points), having little interest in doing things Nearly every day (3 points), feeling tired Nearly every day (3 points), trouble concentrating Several days (1 point), and moving slowly Nearly every day (3 points). This cutoff is often used in research as the baseline for inclusion in behavioral intervention studies.
• Severe Depression (Score 20 to 27): A respondent scoring a 22 is endorsing nearly every symptom at the highest frequency. They are likely reporting feelings of failure, severe sleep disruption, and psychomotor agitation Nearly every day. This score range flags an urgent need for active clinical management, typically involving both pharmacotherapy and psychotherapy.

If you field the PHQ-9, pay special attention to Item 9, which asks about thoughts of self-harm or suicide. If a respondent selects anything other than Not at all for this specific question, standard ethical research protocols require a risk management plan. You cannot simply collect this data passively. Before fielding the survey, you must have an automated system to flag this response and provide the user with immediate crisis resources or a warm handoff to a clinical team.

WHO-5: exactly 5 positive questions

While the PHQ-9 measures the presence of illness, the World Health Organization Five Well-Being Index (WHO-5) measures the presence of psychological health. It is one of the most widely used questionnaires for assessing subjective psychological well-being. Since its introduction in 1998, it has been validated in numerous studies and applied across a wide range of populations.

The defining feature of the WHO-5 is its exclusive use of positive phrasing. Many psychological scales mix positive and negative items, or focus entirely on deficits. The WHO-5 avoids words like "sad," "tired," or "anxious." This makes it highly acceptable to respondents and significantly reduces the friction of fielding mental health questions in non-clinical environments, such as employee engagement surveys or general population research.

Respondents reflect on the past two weeks and rate five statements on a six-point scale from At no time (0) to All of the time (5). The raw score ranges from 0 to 25. To make the score easier to interpret alongside other percentage-based metrics, you multiply the raw score by 4 to get a final index score from 0 to 100.

Here are the 5 positively phrased questions, scoring ranges, and translation tips:

• Question 1: I have felt cheerful and in good spirits.
• Question 2: I have felt calm and relaxed.
• Question 3: I have felt active and vigorous.
• Question 4: I woke up feeling fresh and rested.
• Question 5: My daily life has been filled with things that interest me.
• Scoring ranges: A final multiplied score of 100 represents the highest possible well-being. A raw score below 13 (or a multiplied score below 52) is the established clinical cutoff indicating poor well-being. When a respondent scores below this threshold, it is generally recommended to administer a diagnostic screener like the PHQ-9 to check for underlying depression.
• Translation tips: The WHO-5 has been officially translated into over 30 languages. Never translate the English items yourself using software or informal bilingual staff. Psychological validation requires linguistic validation - ensuring the cultural meaning of "vigorous" or "cheerful" carries the exact same psychometric weight in the target language. Always download the official translated PDFs from the WHO or academic repositories.

Because the WHO-5 is brief and focuses on recent history (the last two weeks), it is highly sensitive to change. This makes it an excellent tool for longitudinal studies. If you are rolling out a new workplace wellness initiative, you can field the WHO-5 before the program starts, and again a month later, to measure actual shifts in subjective well-being rather than just program satisfaction.

NASA-TLX: workload, not preference

The NASA Task Load Index (NASA-TLX) was developed by the Human Performance Group at NASA Ames Research Center in the 1980s. It measures the subjective mental workload a user experiences while completing a specific task.

Workload is distinct from usability. A user might find a software interface highly intuitive and easy to navigate (yielding a high SUS score), but the actual task they are performing - like reconciling a massive financial spreadsheet - might still require intense concentration and effort. The NASA-TLX isolates the cognitive and physical toll of the work itself.

The scale breaks workload down into six distinct dimensions. Respondents rate each dimension on a 21-point scale, typically presented as a visual line anchored by Very Low and Very High.

The original NASA-TLX administration required a two-step process. First, users rated the six dimensions. Second, they completed a series of 15 pairwise comparisons (e.g., choosing whether Mental Demand or Frustration contributed more to the workload). This weighted the final score based on what mattered most to that specific user. Today, most commercial research uses the "Raw TLX" (RTLX) method. This skips the pairwise comparisons entirely and simply averages the six dimension ratings. Decades of subsequent research have shown that the RTLX is just as sensitive and valid as the weighted version for most software and interface testing.

Timing is critical when fielding the NASA-TLX. Because it measures the specific workload of a discrete event, you must administer it immediately after the user completes the task. If you wait until the end of a long, multi-task testing session, the user's memory of the specific cognitive load will blur, and the data will lose its validity.

Reverse-code items (like Rosenberg Q4)

When building long surveys, a major threat to data quality is straight-lining. This happens when a respondent stops reading the questions and simply clicks down the right side of the page, selecting Agree for every item.

To detect and mitigate this, many validated instruments use reverse-coded items. These are questions phrased in a negative direction, meaning that agreement indicates a lower level of the trait being measured. The Rosenberg Self-Esteem Scale (RSES), a widely used 10-item measure of global self-worth, relies heavily on this technique.

For example, Item 1 is positive: I feel that I am a person of worth, at least on an equal plane with others. If a user selects Strongly Agree (4 points), it indicates high self-esteem.

Item 4 is reverse-coded: I wish I could have more respect for myself. If a user selects Strongly Agree for this item, it actually indicates low self-esteem. Before you can sum the total score, you must flip the mathematical value of the answer for Item 4.

If you are using a basic form builder that does not have built-in scoring logic, you will need to handle this conversion in your spreadsheet after exporting the data.

Expert tip: To reverse-score an item in Google Sheets or Excel without writing nested IF statements, use this simple formula: (Maximum Scale Value + Minimum Scale Value) - User Answer.

If your survey uses a 4-point scale where 1 is the minimum and 4 is the maximum, the formula is (4 + 1) - User Answer. If the user selected 4, the math is 5 - 4, resulting in a converted score of 1. If the user selected 1, the math is 5 - 1, resulting in a converted score of 4. This single formula works for any sequential numeric scale.

A common, fatal mistake researchers make is looking at a validated scale, deciding the negative wording is confusing, and rewriting the item to be positive.

Handling a negative validated item

• ❌ Weak: I have high respect for myself. (Rewriting the validated Rosenberg Item 4 to make the data export easier).
• ✅ Strong: I wish I could have more respect for myself. (Keeping the validated wording exactly as published and handling the reverse-coding in the spreadsheet).

Why it works: Altering the wording destroys the psychometric validity of the scale, meaning you can no longer compare your results to established benchmarks.

Quick reference scale directory

Choosing the right instrument depends entirely on what behavior or state you are trying to measure. This directory provides a high-level summary of the five scales covered, helping you match the tool to your specific research objective.

When selecting a scale, consider the burden on the respondent. The total number of items directly impacts completion rates. A 10-item scale like the SUS takes less than a minute to complete, while more complex diagnostic screeners require more time and mental energy.

Before launching any of these tools, run a pilot test with your internal team. Check your data export to ensure the numbers map correctly to the scale values, especially if your survey platform exports text labels (like Strongly Agree) instead of the underlying numeric values required for scoring.

FAQ

Are validated scales free to use for commercial projects?

Many validated scales are entirely free for both academic and commercial use, but the legal mechanisms differ. Some, like the SUS, are functionally in the public domain and can be used without permission. Others, like the PHQ-9, are copyrighted by pharmaceutical companies or universities but are explicitly licensed for free distribution. Always check the original publication or the author's academic institution page to verify the specific commercial use terms before fielding.

How do I properly cite an open access survey scale?

In an internal company report, a simple footnote naming the scale and the original author is usually sufficient. For published white papers or academic research, use standard APA format to cite the seminal journal article that first validated the instrument. Do not link to third-party blog posts that summarize the scale; always direct your citations to the primary source or the official manual.

Can I modify the wording of a validated scale?

No. The moment you change a word, remove an item, or alter the response options (like changing a 5-point scale to a 7-point scale), the instrument is no longer validated. The existing research benchmarks, average scores, and clinical cutoffs will no longer apply to your data. If a question feels slightly outdated, you must weigh the awkwardness of the phrasing against the value of retaining mathematical reliability.

What makes a survey scale scientifically validated?

Validation is a rigorous statistical process, not just a review by experts. A validated scale has undergone psychometric testing to prove reliability (it produces consistent results over time) and validity (it actually measures the construct it claims to measure). This involves techniques like factor analysis to ensure all items point to the same underlying concept, and testing against existing gold-standard measures to prove correlation.

If you are migrating existing validated instruments from old research documents into your digital workflow, you don't have to rebuild the complex scoring matrices and Likert grids by hand. Tools that convert a survey PDF to Google Form via Doc2Form can automatically map established questionnaires directly into your Drive, letting you field rigorous, standardized scales in minutes.