Validated Instruments for an Education Dissertation (2026): Student, Teacher, Classroom and School-Level Measures and How to Report Them

An education dissertation lives or dies on its instruments. A committee will accept a modest sample and a simple design; it will not accept a motivation score from six items the student wrote the night before, or a classroom observation rated by one untrained observer. This guide catalogues the validated instruments that US education dissertations actually use at each level of the system, records what each one measures and who owns it, and sets out what the methods chapter has to report about reliability and validity before a score can be used as a variable. It is organized by the unit you are measuring, because that is how the instrument choice is made.

What committees require of an instrument

Four things, in this order. A published source with psychometric evidence: the development article or manual that reports the factor structure and reliability in a defined sample. Fit to the construct in your theoretical framework: an instrument measures the constructs of the theory it was built from, which is why our guide to choosing a theoretical framework for an education dissertation pairs each theory with its instruments. Permission: some education instruments are free for research use on citation, some require written permission, and some are licensed and charged per administration. And reliability in your own sample, reported per subscale, because a published alpha is evidence about someone else’s participants.

Student-level instruments: motivation, engagement and self-regulation

Two cautions. The MSLQ was developed with college students and is often administered to secondary students without adaptation; if you do that, report the reading level and rerun the factor analysis. And the mindset scale is short enough that alpha will be modest; report it anyway rather than dropping items to raise it.

Teacher-level instruments: efficacy, knowledge and burnout

• Teacher Sense of Efficacy Scale (TSES), Tschannen-Moran and Woolfolk Hoy, 2001. Twenty-four-item long form and twelve-item short form, three subscales (instructional strategies, classroom management, student engagement), nine-point scale. The most cited teacher efficacy measure in US dissertations; free for research use with citation. Its predecessor, the Gibson and Dembo Teacher Efficacy Scale of 1984, still appears in older literature.
• Collective Teacher Beliefs and the Omnibus T-Scale, from the same research group, for school-level efficacy and faculty trust; commonly used in EdD studies of school culture.
• TPACK survey, Schmidt and colleagues, 2009, for preservice teachers, with later versions for in-service teachers by subject. Seven knowledge domains; the instrument is the standard measure in technology integration dissertations, and its factor structure is contested enough that a confirmatory factor analysis in your sample is expected.
• Maslach Burnout Inventory, Educators Survey (MBI-ES). Twenty-two items, three subscales (emotional exhaustion, depersonalization, personal accomplishment). Licensed by Mind Garden and charged per administration; items may not be reproduced in an appendix. Budget for it in the proposal.
• Teachers’ Sense of Efficacy for Literacy Instruction, and similar subject-specific efficacy scales for mathematics and science, when the study is about one content area.

Classroom observation protocols

Observation instruments are where education dissertations most often fail the reliability test, because a protocol scored by one observer with no training is an opinion with a rubric attached.

• Classroom Assessment Scoring System (CLASS), Pianta, La Paro and Hamre, University of Virginia, distributed by Teachstone. Three domains (emotional support, classroom organization, instructional support) scored from cycles of live or video observation; age-banded versions from infant-toddler through secondary, with second-edition Infant-Toddler and Pre-K through third-grade tools now current. Observers are trained and certified through Teachstone, and the certification cost belongs in your budget.
• Danielson Framework for Teaching. Four domains and 22 components, used by many districts for evaluation; usable in research when the district’s trained evaluators produce the scores, but be explicit that evaluation ratings were not collected for research purposes.
• Reformed Teaching Observation Protocol (RTOP), Sawada and colleagues, 2002. Twenty-five items on inquiry-oriented instruction in mathematics and science; free, with a training manual.
• Classroom Observation Protocol for Undergraduate STEM (COPUS), Smith and colleagues, 2013, for two-minute-interval coding of instructor and student behaviors in higher education; free, with short observer training.

Whatever protocol you use, report the number of observers, the training they received, the number of double-coded observations and the agreement statistic (percent agreement is not enough; report Cohen’s kappa or an intraclass correlation).

School-level and system-level instruments

• ED School Climate Surveys (EDSCLS), developed by the National Center for Education Statistics for schools, districts and states, with student (English and Spanish), instructional staff and non-instructional staff versions, covering three domains and thirteen topics, and a survey platform that schools nationwide can access through the Department of Education’s safe and supportive learning site.
• Comprehensive School Climate Inventory (CSCI) from the National School Climate Center, a licensed instrument with student, staff and parent forms.
• Organizational Climate Description Questionnaire and the Organizational Health Inventory, from Hoy and colleagues, for studies of school climate and principal-teacher relations in EdD leadership dissertations.
• Multifactor Leadership Questionnaire (MLQ), Bass and Avolio, for transformational leadership; licensed by Mind Garden.
• Context questionnaires from NAEP, PISA and TIMSS, whose student, teacher and school questionnaires are public, psychometrically documented and free to reuse or adapt with citation, and which make a secondary analysis possible without collecting anything.

Early childhood and higher education measures

Early childhood dissertations lean on the Early Childhood Environment Rating Scale (ECERS-3) for program quality, the Peabody Picture Vocabulary Test and the Woodcock-Johnson batteries for child outcomes (both commercial, with examiner qualifications), and DIBELS or Acadience for early literacy, which are free for schools. Higher education dissertations use the National Survey of Student Engagement (institutional data, not free to individuals), the Institutional Integration Scale from Pascarella and Terenzini for Tinto-framed persistence studies, and the College Self-Efficacy Inventory. Where a scale for a construct genuinely does not exist, the survey-design route applies, and our guide to designing a survey and questionnaire for academic research covers item writing and piloting; adapting a validated scale is almost always the better choice.

Where to find and verify an instrument

Search the Buros Mental Measurements Yearbook for commercial tests and their reviews, PsycTESTS (through the American Psychological Association) for research instruments with permission terms, and ERIC for manuals and development reports. Read the development article rather than a dissertation that used the scale, and check the permission terms on the publisher’s page: Mind Garden instruments (MBI, MLQ, State-Trait Anxiety Inventory) are always licensed; university-hosted instruments (TSES, PALS, SEI) are usually free with citation and sometimes require an email to the author. Record the permission in the methods chapter.

How to present the instruments in the methods chapter

Committees read the instrumentation section as a table before they read it as prose, so give them one. A single table with a row per instrument and columns for construct, instrument name and version, source, number of items and response format, published reliability, reliability in your sample and permission status answers most of the questions a methodologist would otherwise write in the margin. Follow the table with a paragraph per instrument that explains why this measure and not its nearest rival, which is the justification examiners look for and the one students most often leave out. If an instrument was translated, describe the forward and back translation, the reconciliation meeting and the pilot with the target population; if it was shortened, say which items were removed and on what evidence. Put the full instrument in an appendix only when the permission terms allow it, and otherwise include a sample item and the citation to where the committee can obtain the original.

What to report about reliability and validity

Report, for every instrument: its source and the version used; any modifications or translation, with the back-translation procedure; internal consistency in your sample per subscale (Cronbach’s alpha, or McDonald’s omega if the subscale is short or the items are not tau-equivalent); for adapted or newly translated scales, the factor analysis and its fit (CFI and TLI above .90, RMSEA below .08 are the conventions most education committees apply, with the caveat that fit indices are guides rather than rules); inter-rater agreement for observation and rubric data; and the validity evidence you are relying on, in the language of the Standards for Educational and Psychological Testing (content, response process, internal structure, relations to other variables). The wider logic of validity types is covered in our reference on construct, internal and external validity. Once the scores are defensible, the analysis follows the design, and our guide to which statistical test an education dissertation should use takes over from there. For a discipline with parallel conventions on scales and screening ethics, the instruments guide for social work dissertations is a useful comparison.

Frequently asked questions

Can I use a validated instrument without permission in my dissertation?

Only if the developer or publisher has made it free for research use, which is common for university-hosted scales such as the TSES, PALS and SEI and is stated on their pages or in the manual. Licensed instruments such as the MBI and MLQ must be purchased per administration and may not be reproduced in an appendix.

Do I have to report Cronbach’s alpha if the scale is already validated?

Yes. Published reliability describes the developer’s sample, not yours. Report alpha or omega for each subscale in your own data, and comment on any subscale that falls below .70.

Can I modify items in a validated scale?

You can, but the scale is then adapted rather than validated, and you must report the changes, pilot the adapted version and run a factor analysis to show the structure held. Wording changes to fit a subject or grade are common; dropping subscales or rewriting constructs is a new instrument.

How many observations does a classroom observation study need?

Enough cycles per teacher to produce stable scores, which for CLASS is typically four cycles per observation and several observations per teacher, and enough double-coded observations, commonly 20 percent, to estimate inter-rater reliability.

Is a district teacher evaluation rating usable as a research variable?

Often, with caveats: the ratings were produced for evaluation, by evaluators whose training and agreement you may not be able to document, and their distributions are typically compressed. Report those limitations and, where possible, triangulate with another measure.

Where can I read the actual items of an instrument before choosing it?

In the development article or manual (ERIC hosts many, including the MSLQ manual), in PsycTESTS, or on the developer’s university page. If you cannot see the items before committing, treat that as a reason to look for another instrument.