New webinar on 3rd March 2026 | 9:00 am PT
In this webinar, discover how Sopact Sense revolutionizes data collection and analysis.
What a longitudinal study is, the most famous ones (Harvard Grant, Framingham, Dunedin), the four types, and the advantages and disadvantages textbooks miss.
It is how we know what we know about how lives unfold. Happiness across eighty years. Heart disease across three generations. Development from infancy through middle age. A single-moment study cannot answer those questions. A longitudinal study is the design that can.
The most cited longitudinal studies have shaped what we know about human life. The Harvard Grant Study has followed the same men since 1938. The Framingham Heart Study has tracked the same cardiovascular cohort since 1948. The Up Series has revisited the same fourteen British seven-year-olds every seven years since 1964. The Dunedin Study has tracked 1,037 New Zealanders born in 1972 through to middle age. The English Longitudinal Study of Ageing has followed adults aged fifty and above since 2002.
This guide explains what makes these studies longitudinal, the four types of longitudinal study you can run, the practical advantages and disadvantages textbooks list, and the operational work behind every multi-decade study. No prior research background is needed. Examples come from psychology, public health, and applied program evaluation.
A longitudinal study connects each person's measurements across waves. The Dunedin Study tracks the same 1,037 New Zealanders born in 1972 through to middle age. The connection across waves is what makes it a longitudinal study, not the length.
Each bar below is a longitudinal study still running today. The earliest started before World War II. The most recent started in this century. Each one made findings that no single-moment study could have made. The pattern that connects them is the same: the same units, observed across time, with each unit's measurements connected wave by wave.
Started 1938 . 268 Harvard sophomores . Happiness, health, and what predicts a meaningful life.
Started 1948 . 5,209 residents and now their grandchildren . The evidence base for modern cardiovascular medicine.
Started 1964 . 14 British seven-year-olds . Filmed every seven years, now in their sixties.
Started 1972 . 1,037 New Zealanders born that year . Multidisciplinary tracking from infancy into middle age.
Started 2002 . Over 12,000 adults aged 50 and above . Health, economics, and aging policy in the UK.
These five studies represent decades of psychology, public health, and aging research. Each one is run by a research institution with dedicated infrastructure: study coordinators, custom databases, decades of institutional continuity. The structural choices behind them (tracking the same people, locking the survey wording, planning for attrition) are the same choices any applied longitudinal study has to make, on a shorter timeline and with a smaller team.
The terms longitudinal study, longitudinal research, longitudinal design, and panel study are often used as if they meant the same thing. They mostly do, with small shades of difference. The five answers below cover the five question forms that send readers to this page.
A longitudinal study is research that surveys or observes the same people, organizations, or units more than once across time, and connects each unit's answers from one round to the next. Each round is called a wave. The defining feature is that the same units appear at every wave, so changes within the same unit can be measured directly rather than inferred from comparing different groups.
The Harvard Grant Study has tracked the same men since 1938. The Framingham Heart Study has followed the same families since 1948. Both are longitudinal studies because the same participants are observed across waves, not because they are long.
The standard textbook definition of a longitudinal study: a research design in which the same individuals are observed or tested repeatedly at different points in their lives. Some textbooks add the requirement that the time span be substantial (months or years rather than days), but no fixed minimum exists. The defining property is the repeated observation of the same units, not the length of time over which the observation happens.
In APA-style methods writing, the term longitudinal design is preferred when the focus is on planning, and longitudinal study when the focus is on the work being done. Outside of methodological writing, the two are usually interchangeable.
The word longitudinal comes from "longitude," meaning length. A longitudinal study has length in time: the study is stretched across multiple time points rather than compressed into one. Saying a study is longitudinal does not commit you to a length, only to the structure. A two-wave study six weeks apart can be longitudinal. A multi-decade study with annual waves can also be longitudinal.
What makes the structure work is the connection between waves. Two surveys six weeks apart with no way to match the same person's answers between them is not a longitudinal study; it is two cross-sectional studies. Length without connection is not longitudinal data.
In psychology, a longitudinal study follows the same participants across months, years, or decades to measure how their behavior, cognition, personality, or development changes over time. Three iconic examples come from psychology: the Up Series followed fourteen British children from age seven into their sixties, the Dunedin Study tracked 1,037 New Zealanders from infancy through middle age, and the Harvard Grant Study followed Harvard sophomores into their nineties.
Psychology textbooks usually distinguish longitudinal studies from cross-sectional studies and from cross-sequential designs. The cross-sequential design combines the two: several cohorts followed longitudinally so that age effects and cohort effects can be separated. For most coursework, the simple definition holds: longitudinal means same people, more than once, across time.
The cleanest example to remember: imagine surveying 320 workforce-training participants at intake, at the end of training six months later, then twelve and twenty-four months after exit. The same participants answer at every wave, and each person's answers are connected through a tracking ID set at the start. By the end of month twenty-four the program can report that wages rose for 184 of the 240 participants who responded at twelve months, not only that the group average rose. That within-person measurement is what a longitudinal study produces.
For a deeper academic example, see the Dunedin Study walked through in section five below. For more cluster-level detail on the structural concepts, see the longitudinal design hub.
These four designs share features with longitudinal study and are often used in the same conversations. Each one differs from longitudinal study in a specific way. Knowing the difference is what tells you whether the research you are reading or running is what you think it is.
A cross-sectional study compares different people at a single point in time. It can show how groups differ but not how individuals change. Comparing today's twenty-year-olds to today's sixty-year-olds is cross-sectional. Following the same people from age twenty to age sixty is longitudinal. The first cannot tell you whether anyone changed.
A case study examines one unit (a person, an organization, an event) in depth. Case studies are sometimes longitudinal (tracking the unit across time) and sometimes not. The defining feature is the focus on one unit, not the time dimension. A longitudinal case study is one unit followed across time; most case studies in management research are not longitudinal.
A cohort study follows a group defined by a shared starting point: a birth year, a school entry year, a program enrollment quarter. A cohort study is a kind of longitudinal study (or can be), distinguished by how the group is defined rather than by what is done with the group. The Framingham Heart Study is a cohort study because the cohort was defined by living in Framingham in 1948.
An experiment assigns participants to conditions (treatment vs control) and measures outcomes. Experiments can be longitudinal (measurements at multiple waves) or cross-sectional. The defining feature is random assignment to conditions, not the time dimension. Most longitudinal studies are observational: the researcher follows what happens but does not assign treatments.
The textbook definition of longitudinal study is technically correct and useful for an exam. The six items below are what readers of those textbooks notice once they try to read or run a longitudinal study in real life. None of these are in the formal definition. All of them shape what longitudinal data can and cannot tell you.
Identity across waves is the structural requirement.
A longitudinal study is not a series of surveys with the same questions. It is a series of surveys connected to the same participants. Two waves with different people in them is a trend study or two cross-sectional studies, not a panel longitudinal study. The connection across waves is what makes the data longitudinal.
What this changes: the analysis can show within-person change, which is the whole point of running a longitudinal study.
Each participant is their own comparison.
The deepest reason to run a longitudinal study is that each participant becomes their own comparison. Wave 1 Maria is the right baseline for Wave 5 Maria. The within-person change controls for everything stable about Maria (her background, her personality, her circumstances) that a between-group comparison cannot control for.
What this changes: longitudinal studies need much smaller samples to detect a real effect than cross-sectional studies do.
There is no fixed minimum.
A two-wave study six weeks apart is longitudinal if six weeks is long enough for the outcome to change. A multi-decade study is longitudinal if the change being measured takes decades. Length is not a defining feature; it is a consequence of the outcome being measured. The Up Series uses seven-year intervals because that is what the filmmakers chose, not because seven years is the right interval.
What this changes: a study can be longitudinal without being long. Pre-and-post is a two-wave longitudinal design.
The people who drop out are systematically different.
Across waves, participants drop out. The people who drop out are usually different from the people who stay: less engaged, harder to reach, in worse circumstances. If attrition is ignored, the conclusions describe the people who survived rather than the people who started. Most longitudinal studies need an attrition analysis showing whether dropouts differ from completers on baseline measures.
What this changes: the size of the conclusions you can draw shrinks as attrition rises, even if statistical power is fine.
The work behind the data is what kept them running.
Harvard Grant has a study coordinator who has been with the project for decades. Framingham has a building, a clinic, a database, and three generations of staff. The Up Series has a single director who returns every seven years. The studies are famous because the data is good; the data is good because someone built the infrastructure to keep finding the same people for decades.
What this changes: any applied longitudinal study needs the same kind of infrastructure on a smaller scale.
Quantitative and qualitative usually run together.
Most major longitudinal studies collect both kinds of data. Framingham takes physical measurements and structured interviews. Dunedin combines health assessments with detailed life-history interviews. The qualitative data explains patterns the quantitative data shows. A longitudinal study that collects only numbers is choosing to leave half the analysis on the table.
What this changes: the survey tool needs to handle open-ended responses across waves the same way it handles structured fields.
A longitudinal study can take several shapes. The four most common types (panel, cohort, prospective, retrospective) are not exclusive; one study can be a prospective panel cohort, for example. Each row below is a planning decision the researcher faces. The "broken way" is the version that loses the structural advantage of running a longitudinal study in the first place. The "working way" is the version that survives to publication.
Panel design vs trend design
Sample new participants at each wave from the same general population, then claim the data is longitudinal because the population is stable. The connections between waves do not exist; only the population does.
Decide before Wave 1 whether the study is a panel design (same individuals every wave) or a trend design (different individuals from the same population). Each answers a different question. Both are legitimate; mixing them mid-study is not.
Whether you can answer "how did this person change?" or only "how did this group change?"
Prospective vs retrospective
Mix prospective and retrospective data without flagging which is which. Some Wave 1 measurements are from records collected at the time; others are reconstructed from interviews months or years later. The two are treated as the same kind of data.
Choose prospective when accuracy matters more than speed. Choose retrospective when the outcome already happened and forward collection is impossible. Document which wave is which kind of data, and account for recall error in the analysis.
Whether your answers are based on records collected at the time or memories reconstructed later.
Two vs three or more
Run only two waves on the assumption that pre-and-post is enough, then realize at analysis that two waves cannot show a trajectory. You know the start and the end; you do not know whether the change was early, late, gradual, or interrupted.
Match wave count to the trajectory shape you are looking for. Two waves answer "did it change." Three to five answer "when did it change and how fast." More than five let you fit a trajectory and group people by trajectory shape.
Whether you can detect early-vs-late change patterns or only endpoints.
Set by deadline vs set by outcome
Wave timing is set by the funder's reporting cycle. The end-of-quarter report needs data, so a wave runs whether or not enough time has passed for the outcome to change. The data captures noise rather than signal.
Set wave timing by the outcome being measured. Skill change in weeks needs short intervals. Wage change in months needs medium intervals. Health outcomes in years need long intervals. The funder's report is built around the wave schedule, not the other way around.
Whether the data captures real change or measurement noise.
After the fact vs before Wave 1
Sample size is set for Wave 1 and the team plans to "see what happens" with attrition. By the final wave, half the original sample is gone, the analysis is underpowered, and the team cannot tell whether the people who stayed are representative of the people who started.
Sample size is set for the final wave with a realistic attrition estimate. To analyze 200 people at Wave 5, plan for 320 to 380 at Wave 1. Build attrition analysis into the protocol so dropout patterns can be reported alongside the main results.
Whether the final wave can be analyzed at all.
Numbers vs numbers plus narrative
Collect only structured survey data because it is faster to analyze. The numbers show that something changed, but no record exists of why. The team writes the discussion section by inferring causes from the literature rather than from the participants.
Add open-ended response or short narrative to at least the key waves. The qualitative data explains patterns the quantitative data shows: who changed, in what direction, and what they say happened. Mixed-method longitudinal is what the major studies do.
Whether you can explain why the change happened, not only that it did.
The six choices above shape each other. A two-wave panel study with quarterly intervals and no attrition plan is a different design from a five-wave cohort study with outcome-driven intervals. Either can be the right design for the right question. Neither is the right design for every question. The first decision (same individuals or same population) is the one that closes off the most options for everything else.
The Dunedin Multidisciplinary Health and Development Study has tracked 1,037 New Zealanders born in 1972 from infancy into middle age. It is the most cited longitudinal study of its generation. The structural choices the Dunedin team made are the same choices any longitudinal study has to make, scaled to a different timeline.
We started in 1972 with 1,037 babies born in Dunedin. We have followed them through childhood, adolescence, and into their early fifties. Fifty-two years later, we still see ninety-six percent of the original cohort at every assessment wave. That retention rate is not luck. It is dedicated coordinators who know each participant by name, a single tracking record that has lived with each person their whole life, and a research design that built in the trust the study would need at year forty before Wave 1 ever happened.
Paraphrased from public statements by Dunedin Study leadership, 2022
Linked at collection, every wave
A dedicated study coordinator team
A small team that has been with the project for decades. Each coordinator knows specific participants by name and tracks their address changes between waves.
A custom database, maintained continuously
One record per participant, growing across decades. Every wave, every measurement, every interview filed against the same ID. The database has outlived several generations of statistical software.
A research clinic in Dunedin
Participants come to a familiar location every assessment wave. The clinic is a single physical anchor that has not moved in fifty years.
University funding across decades
Multi-decade research funding from the New Zealand health system and international research grants. The funding model assumes the study will keep running.
A few program staff, not a study team
Two or three people running collection alongside their other duties. No one whose only job is participant tracking.
A survey tool, not a custom database
An off-the-shelf survey platform that produces a separate file per wave. The connection between waves is built at analysis time, in spreadsheets, by hand.
A short timeline, not five decades
Twelve to thirty-six months, three to six waves, an outcome that has to be reported to a funder by a fixed date.
A program budget, not a research grant
The longitudinal study is one part of the program's measurement work, not the program itself. Money for dedicated study coordinators is not in scope.
Dunedin has fifty years and a research clinic. An applied program has eighteen months and a survey tool. The structural choices are the same: a tracking ID set at first contact, locked survey wording across waves, a wave schedule matched to the outcome, and a plan for attrition before Wave 1. What Dunedin's coordinators do by hand across decades, an applied team has to do through software across months. Sopact Sense is built to give applied longitudinal work the same structural reliability without the university overhead.
Most longitudinal studies sit in one of three settings. The first two are where the famous studies live, run by research universities and dedicated clinics across decades. The third is where most modern applied work happens, in program evaluations on shorter timelines with smaller teams. The structural choices are the same in all three; the infrastructure that supports those choices is different.
University-led. Multi-decade. Birth cohort or school-entry cohort.
Typical shapeAn academic research team at a university recruits a birth cohort or school-entry cohort numbering in the hundreds to low thousands. The waves run every two to seven years for the lifespan of the participants. The infrastructure is built into the institution: a dedicated coordinator team, a custom database, multi-decade funding lines, and a research clinic that participants return to. The Up Series, Dunedin, and most major birth-cohort studies fit this shape.
What breaksThe biggest threats are funding interruptions across decades and the slow loss of institutional memory as the original investigators retire. Participants who emigrate become hard to follow without international research partnerships. Wave intervals set by traditional convention (every five years, every seven years) can be too sparse to catch fast-moving developmental change.
What worksContinuous coordinator-participant relationships across decades. A single tracking ID assigned at recruitment that lives in the database for life. Survey wording locked at Wave 1 and only added to (never replaced) at later waves. Mixed-method data collection (structured measurements plus life-history interviews) at every wave so the qualitative axis stays connected to the quantitative axis.
The Dunedin Study has retained 96 percent of its original 1,037 participants across 52 years. That retention rate is the result of dedicated infrastructure, not luck. Most academic longitudinal studies aim for and report attrition figures from this benchmark.
Clinic-led. Multi-decade. Population cohort, often regional.
Typical shapeA health research institution recruits a regional cohort of several thousand to tens of thousands of adults. Waves run every two to four years, with biomarker assessments alongside structured surveys. The cohort is often expanded across generations as the original participants age and their children enroll. Framingham, ELSA, the Whitehall Study, and the Nurses' Health Study all fit this shape.
What breaksLoss of trust in research institutions can drop response rates between waves. Participants who relocate out of the catchment region become expensive to follow. Diagnostic categories drift across decades (a mental-health diagnosis named one thing in 1970 may be defined differently now), making the survey-wording lock harder to maintain than it sounds.
What worksA physical clinic anchor that participants return to. Standardized biomarker collection that produces measurements directly comparable across waves regardless of survey-question changes. A clinical-data backbone that holds the participant record and a separate survey backbone for self-report data, both linked by the same tracking ID.
The Framingham Heart Study has run since 1948 and now follows three generations of the original 5,209 residents. Most of what cardiovascular medicine knows about heart-disease risk factors comes from this single longitudinal cohort.
Program-led. Months to years. Cohort or rolling enrollment.
Typical shapeA program team running workforce training, education, public health, or impact-fund work tracks participant outcomes across waves. Cohort sizes are usually a few hundred to a few thousand. Waves run every three to six months for the duration of the program plus six to twenty-four months of follow-up. The team is small (two or three people running measurement alongside their other duties), and the timeline is short relative to academic studies but long relative to single-survey work.
What breaksOff-the-shelf survey tools produce a separate file per wave with no built-in connection. The connection has to be built at analysis time, in spreadsheets, by hand. Email addresses change as participants change jobs. Names get misspelled. Twenty to forty percent of records fail to match across waves. The team accepts a partial match and reports group averages instead of within-person change. The structural advantage of running a longitudinal study is lost in spreadsheet matching.
What worksA purpose-built data collection tool that holds one record per participant and grows the record across waves. A tracking ID set at first contact and stored against the participant, surviving any later change to email, phone, or last name. Real-time visibility into who is late inside a wave's response window versus who is lost. The matching work happens at collection time, when it can still be fixed, rather than at analysis time, when it cannot.
A workforce-training cohort of 320 participants tracked across 24 months can produce within-person wage-change data on 240 of the 320 (75 percent retention) when the tracking ID is set at intake. The same cohort produces only group averages when the matching work is deferred to analysis.
The famous longitudinal studies are run by research institutions with custom databases, dedicated coordinator teams, and decades of institutional continuity. The off-the-shelf survey tools that most applied teams use (Google Forms, SurveyMonkey, Qualtrics, Typeform) were built for the opposite kind of work: a single survey, fielded once, results exported once. They collect well. The structural gap is what happens after collection across multiple waves: the same person's answers need to be connected by a tracking ID set at first contact, and most survey tools produce a separate file per wave with no built-in way to make that connection. The matching work that a research-university coordinator does by hand becomes spreadsheet work for the applied team.
Sopact Sense is built around the structure that academic longitudinal studies developed over decades, applied to the timeline and team size of program-evaluation work. Each participant is one record that grows across waves. The tracking ID is set at first contact and survives any later change to email, phone, or name. Partial responses stay attached to the participant rather than becoming orphan rows. The matching that other tools defer to analysis time happens here at collection time, which is the only point at which it can still be fixed. The result is research-university structural reliability without the research-university overhead.
Definitional questions, comparison questions, and execution questions. Each answer is short on purpose. The fuller treatment is in the relevant section above.
A longitudinal study is research that follows the same people, organizations, or units across multiple points in time. Each round of measurement is called a wave. Because the same units are observed at every wave, the study can show how individuals change rather than only how groups differ. The Harvard Grant Study has followed the same men for over eighty years. The Framingham Heart Study has tracked the same cardiovascular cohort across three generations. Both are longitudinal studies.
In psychology, a longitudinal study follows the same participants across months, years, or decades to measure how their behavior, cognition, personality, or development changes over time. The Up Series followed fourteen British children every seven years from age seven into their sixties. The Dunedin Study has tracked 1,037 New Zealanders born in 1972 through to middle age. These studies are what allow psychology to make claims about development across the lifespan that no single-moment study could make.
The most cited examples include the Harvard Grant Study (started 1938, 268 Harvard sophomores tracked into their nineties), the Framingham Heart Study (started 1948, 5,209 residents and now their grandchildren), the Up Series (started 1964, fourteen British seven-year-olds revisited every seven years), the Dunedin Multidisciplinary Health and Development Study (started 1972, 1,037 New Zealanders), and the English Longitudinal Study of Ageing (started 2002, over 12,000 adults aged fifty and above). Each one made findings that no cross-sectional study could have made.
A longitudinal study follows the same people across time. A cross-sectional study compares different people at one moment in time. The first answers questions about how individuals change. The second answers questions about how groups differ at one moment. A study comparing twenty-year-olds to sixty-year-olds today is cross-sectional; a study following the same people from age twenty to age sixty is longitudinal.
Longitudinal studies measure within-person change directly. They can show whether the same individuals improved, worsened, or stayed flat. They can sometimes establish that one event preceded another, which strengthens causal interpretation. And they can describe trajectories that cross-sectional studies cannot see, such as people who improve early then plateau versus people who improve late. Without longitudinal evidence, claims about how people change across time are mostly inference.
Longitudinal studies cost more and take longer than cross-sectional studies. Participants drop out across waves, and the people who drop out are rarely random. Survey questions written years ago may no longer be the right questions. Practice effects can change what the answers mean if the same survey is given too often. And the operational work of finding the same people at each wave, especially across decades, is the most common reason longitudinal studies fail to deliver clean data.
Long enough that the change being studied has time to occur. Two waves six weeks apart can be a longitudinal study if six weeks is long enough for the outcome to change. Multi-decade studies are common in developmental psychology and aging research because the changes those studies care about take decades to unfold. The minimum length is set by the outcome, not by a fixed rule.
It can be either or both. The defining feature of a longitudinal study is that the same units are observed across time, not the type of data collected. Quantitative longitudinal studies use scales, counts, and structured surveys at each wave. Qualitative longitudinal studies use repeated interviews or observations. Mixed-method longitudinal studies combine both, often using the qualitative data to explain patterns the quantitative data shows.
Four types are common. Panel studies survey the same individuals at every wave. Cohort studies follow a group defined by a shared starting point, such as a birth year or program cohort. Prospective studies plan their waves at the start and collect data forward in time. Retrospective studies ask current participants about past time points. The four types are not exclusive: a panel cohort study can be either prospective or retrospective.
A panel study is the strongest form of longitudinal study. The same individuals are surveyed at every wave, and each person's answers across waves are connected through a tracking ID. Panel studies allow within-person change to be measured directly because each person serves as their own comparison. The British Household Panel Survey, the German Socio-Economic Panel, and the Panel Study of Income Dynamics are major examples in social science.
A prospective longitudinal study plans its waves at the start and collects data forward in time. Each measurement is taken at the moment it happens rather than reconstructed from memory later. Prospective studies are more accurate than retrospective studies but cost more because they take real time to run. Most major cohort studies in health and developmental psychology are prospective.
A retrospective longitudinal study asks current participants about past time points. The researcher works backward through the participant's history rather than forward through real time. Retrospective studies are faster and cheaper than prospective studies but more prone to recall error because participants are reconstructing the past from memory. Retrospective designs are often used when prospective data collection is impossible, such as when studying a rare outcome that has already occurred.
A longitudinal study can establish that one event preceded another in time, which is one of the conditions for causation but not the whole condition. Without random assignment, a longitudinal study cannot rule out that a third factor caused both events. The Framingham Heart Study established that smoking precedes cardiovascular disease in time and that the relationship is dose-responsive, but the strongest causal claims still come from combining longitudinal evidence with experimental and mechanistic studies.
Famous longitudinal studies are run by research universities with dedicated infrastructure: custom databases, study coordinators, and decades of institutional continuity. Most program evaluations and applied longitudinal studies cannot afford that. Sopact Sense is built to give applied longitudinal work the same structural reliability without the university overhead: each participant is one record that grows across waves, the tracking ID is set at first contact and survives across changes to email or phone, and partial responses stay attached to the participant rather than becoming orphan rows.
Yes, and most longitudinal work outside of academia happens this way. Workforce-training programs, education foundations, public-health initiatives, and impact funds run longitudinal studies on cohorts of a few hundred to a few thousand participants. The structural choices are the same as in academic studies (tracking ID at Wave 1, stable wording across waves, attrition planning) but the timeline is shorter (months to a few years rather than decades) and the team is smaller. The work that an academic study coordinator does manually is the work that purpose-built software does for an applied team.
Six pages explain the longitudinal cluster from different angles. The hub explains the architecture. The comparison page sits between longitudinal and cross-sectional. Three sibling pages cover the data format, the analysis, and the collection tools. The simplest case (pre-and-post) is its own page.
The structural concept. How a longitudinal study is built wave by wave, and what the architecture has to do for the data to hold.
When each design is the right one. What each can answer and what each cannot. The trade-offs textbooks summarize and what they leave out.
What longitudinal data actually looks like once it is collected. Long format, wide format, and why the choice matters before analysis starts.
How within-person change is modeled. Growth curves, mixed-effects models, and the statistical methods that make longitudinal data answer questions cross-sectional data cannot.
The collection-tool angle. What a longitudinal survey requires that a one-shot survey does not, and what to look for when choosing the tool.
The two-wave longitudinal study. What pre-post can answer, what it cannot, and the matching work that is the same in pre-post as in any longer study.
A sixty-minute methodological conversation. Bring the question you are trying to answer. Together we look at whether longitudinal is the right design, what wave count and intervals would fit, and what the operational work would look like across the timeline you have. No pitch deck. No commitment.
Unmesh Sheth
Founder and CEO, Sopact
