MPH-507 · Week 4 of 8 · Goals, objectives and the logic model

MPH-507 Week 4 Objectives and the Logic Model: How to Write It

The short answer

MPH-507 Week 4 is the structural centre of the course. Objectives written with a number, a population, a direction and a deadline are what everything downstream measures against, and the logic model is the diagram that proves the chain from inputs to outcomes has no missing link. Deliverables at this stage are graded on whether a reader could declare each objective met or missed on a given date, which is a harder test than it sounds. Your section may print this as MPH 507 or MPH507; it is the same course. Chamberlain publishes no syllabi outside Canvas. The placement here is our teaching judgment from the course's catalog arc; your section's rubric decides what your week actually asks.

MPH-507 Week 4 grading scale at Chamberlain, the criterion levels this assessment is scored on, from Chamberlain Tutors
How Chamberlain grades MPH-507 Week 4, visualized by Chamberlain Tutors.

What an auditable objective set delivers

A program audit that goes badly usually goes badly in a specific way. The auditor asks what the program committed to, reads the objectives, and finds that none of them can be tested. Improve access. Increase awareness. Strengthen partnerships. The program may have done excellent work, and the record cannot show it, because nobody wrote down what success would look like in advance. The reverse also happens: a program hits every target and the auditor notes that the targets were set below the level already achieved at baseline. Both failures are authored at this stage, in the objective section, months before anyone collects data.

The territory is the difference between a goal and an objective, and the tiering of objectives by time horizon. A goal is a single sentence naming the long-term change, without numbers. Objectives are testable commitments, and they come in tiers with different clocks. Short-term objectives target what changes first, typically knowledge, skill, intention, policy adoption or service availability. Intermediate objectives target behavior and service use. Long-term objectives target the health status change that motivated the work and may take years, which is why they are stated with reference to the literature connecting the intermediate change to them rather than promised inside a grant period.

Each objective needs five components on the page: the population, the direction and size of change, the indicator, the baseline it moves from, and the date by which it will be measured. The baseline is the component most often missing, and its absence is fatal, because an objective to reach a stated coverage level means nothing until a reader knows whether coverage was already there.

The logic model is the other half. Inputs, activities, outputs, short and intermediate outcomes, and long-term impact, arranged so each column is produced by the one before it. Its value is diagnostic: a column you cannot fill is a gap in the plan, and an outcome with no activity behind it is a promise nothing in your program produces. Deliverables here are usually an objective set with a logic model, sometimes a narrative walkthrough. Canvas posts do not reopen after submission, so put the number in the objective before you post it.

A method for writing objectives that survive audit

Six moves that produce a testable commitment set.

  1. Write the goal once, in one sentence, without numbers

    Name the long-term change in the population you defined. If your goal contains a list of activities, it is a work plan rather than a goal, and every objective under it will inherit the confusion.

  2. Establish each baseline before writing the target

    Find the current value of the indicator in your population, with its source and year. Objectives written without a baseline either restate the status quo as an achievement or commit to a change nobody can size.

  3. Size the change from evidence rather than ambition

    Look at what comparable programs achieved in comparable populations and set your target inside that range. A target far above anything documented is a rubric problem now and a credibility problem when the evaluation reports.

  4. Assemble the five components in one sentence each

    Population, direction, magnitude, indicator, baseline and deadline. Write them in a fixed order across all objectives so a reader can scan for a missing component, and so you can find it yourself.

  5. Tier the set by time horizon and say which tier you will measure

    Short-term, intermediate and long-term, with the evaluation aimed at the first two and the third referenced to literature. Programs that promise to measure health status inside a one-year window are not credible and the feasibility row notices.

  6. Fill the logic model and repair what will not fill

    Work across the columns and stop at the first cell you cannot complete. That empty cell is the real finding of the exercise. Add the activity, drop the outcome, or state plainly that the link depends on a factor outside your control.

A layout and word budget for objectives and a logic model

Our frame for this deliverable, sized for roughly 1,200 to 1,500 words of narrative alongside the model. It is our own outline rather than anything the university issues, and your section's scoring guide outranks it wherever the two disagree. The logic model itself and an objective table will together carry perhaps 300 words of what would otherwise be prose, and that budget belongs in the assumptions and measurement paragraphs.

SectionWhat belongs in itWord target
Goal statementOne sentence naming the long-term change, followed by two sentences tying it to the determinant you targeted.90 to 120
BaselinesThe current value of each indicator with its source, year and geography, and a note where a baseline had to be estimated.180 to 220
Objectives, tieredShort, intermediate and long-term objectives, each carrying all five components in a consistent order.280 to 340
Target justificationWhy each magnitude is achievable, referenced to what comparable programs have produced.200 to 250
Logic model walkthroughThe chain read across in prose, naming what each column contributes and where the weakest link sits.230 to 290
Assumptions and external factorsWhat has to be true for the chain to hold, and which conditions outside your control could break it.180 to 230

Evidence craft for targets and chains

Cite the source and year of every baseline value. A baseline is the number your program will be judged against, so it needs the same provenance as any other evidence: the system it came from, the geography it covers and when it was collected. Where you had to estimate a baseline, say how, because an estimated baseline with visible method is defensible and an unexplained one is not.

Justify target magnitudes with comparable results. When you commit to a change of a stated size, cite where a similar intervention produced a similar change, with the population and design in front of the result. Ten words of provenance turns an arbitrary target into an evidence-based one, and the row scoring objectives is usually looking for exactly that.

Keep outputs and outcomes in separate columns and separate sentences. Sessions delivered, materials distributed and people reached are outputs, which measure delivery. Changes in knowledge, behavior, service use or health status are outcomes. Reporting an output where an outcome belongs is the single most common structural error in program writing and it is immediately visible in a logic model.

State the assumptions between columns explicitly. Every arrow in a logic model carries an assumption: that people reached will attend, that attending changes intention, that intention changes behavior, that behavior changes the outcome. Naming the two or three most fragile assumptions, and saying what evidence supports them, is the paragraph that most reliably lifts this deliverable out of the middle band.

Five mistakes that cost points in this week's territory

  • Adverbs where numbers belong. Significantly, substantially and effectively are the three words that turn an objective back into a hope, and they are easy to find before submission.
  • No baseline. A target with nothing to move from cannot be evaluated, and it may already have been met before the program started.
  • Health status promised inside a program year. Long-term outcomes on a short clock are the clearest signal that the objectives were written without a time structure.
  • Outputs presented as outcomes. Four hundred people attended is a delivery fact and says nothing about whether anything changed in them.
  • A logic model that nobody references. A diagram inserted and never discussed earns nothing, and a diagram whose columns do not match the objectives above it is worse than none.

Before you submit

  • The goal is one sentence and contains no activities
  • Every objective carries population, direction, magnitude, indicator, baseline and date
  • Each baseline has a source, a year and a geography
  • Target sizes are justified against documented comparable results
  • Objectives are tiered and the evaluation targets the tiers it can reach
  • The logic model is referenced in the prose and matches the objective set

Writing objectives and a logic model?

Send the prompt, the scoring guide and the determinant you chose to target. A premium original draft comes back in 24 to 48 hours with baselines sourced, targets justified and every column of the model filled, and revisions run until the grade lands.

Questions students ask about this stage

How many objectives should a plan have?
Fewer than most students write. A program of the size a course deliverable describes can usually support one or two short-term objectives, one or two intermediate, and one long-term, which is four to five total. Every objective creates an obligation later: an indicator, a data source, a collection point and an analysis, all of which have to appear in your evaluation section and be paid for in your budget. Students who write nine objectives in Week 4 spend Week 7 discovering that they have committed to a measurement program larger than the intervention. There is also a scoring effect. Four objectives written completely, each with all five components and a justified target, score better than nine written loosely, because the rows are assessing quality of specification rather than quantity of ambition.
What do I do when there is genuinely no baseline for my indicator?
You have three defensible options and one that is not. You can select a different indicator that does have a baseline, which is often the right call when the substitute measures nearly the same thing. You can establish the baseline as the first activity of the program, writing an objective that commits to measuring it by a stated date and then setting the target afterward, which is honest and common practice. Or you can construct an estimated baseline from the closest available proxy, stating the proxy, the adjustment and the uncertainty in the sentence. What does not work is writing a target with no reference point and hoping nobody asks, because the evaluation section two stages later will have nothing to compare against and the gap will be visible in a document you have to keep building on.
Does the logic model go before or after the objectives?
In the document, the objectives usually come first and the model follows, because a reader needs to know what you are committing to before they can read a diagram of how it happens. Check your prompt, since some ask for a specific order or a specific model format. In your drafting process, however, build the model first and in dependency order. Sketch the columns, find the empty cells, then write the objectives against the outcome columns that survived. Working that way prevents the common failure where an objective and the model disagree, which reviewers spot instantly and which suggests the two sections were written on different days without reconciliation. Whichever order you present them in, walk the model through in prose, because a figure that is never mentioned in the text is treated as decoration.

Keep going

Online now