Compiled by: Kevin Rotich
Research is a planned, systematic investigation that answers a defined question by collecting and analysing evidence. A laboratory technician meets research in many forms: validating a new method before it replaces an old one, finding why a batch failed a quality test, comparing two treatments, monitoring an environmental indicator over a season, or developing a product. In each case the work is research, and not routine testing, when its aim is to produce knowledge that was not available before and when its steps can be inspected and repeated by someone else. Research methods are the techniques used to collect and analyse evidence, and the study of why one technique suits a question better than another is research methodology.
Research follows a cycle of stages, each of which is taught in this unit.
Stages 1 to 6 are written as the research proposal, which is why this chapter, on preparing for research, leads to proposal writing.
A method is a specific technique: a titration, a membrane filtration count, an interview guide or a t-test. Methodology is the reasoning that links the problem to the choice of techniques and justifies it. A proposal states the methodology (for example, a controlled laboratory experiment with replicated treatments) and then lists the methods (the standard procedures, the instruments and the statistical tests).
| Type | Purpose | Laboratory example |
|---|---|---|
| Basic (pure) | Extend understanding with no immediate use in mind | Studying how temperature changes the rate of a reaction in a new solvent |
| Applied | Solve a practical problem | Testing whether a cheaper reagent gives the same result as the one in use |
| Exploratory | Learn about a topic that is little understood so that later questions can be set | Interviewing technicians to find out what causes repeat testing |
| Descriptive | Describe the characteristics of a situation or population | Measuring the range of pH in water from a set of springs |
| Explanatory | Explain why something happens or what causes it | Testing whether the type of bottle changes the measured pH of stored water samples over one week |
| Evaluative | Judge how well a method, product or programme works | Comparing the performance of two disinfectants on bench surfaces |
| Action research | Improve practice in a workplace through repeated cycles of planning, acting and reviewing | A laboratory trying a new sample-labelling system and adjusting it over three months |
| Feature | Quantitative | Qualitative | Mixed |
|---|---|---|---|
| Data | Numbers from measurement and counts | Words, images and observations | Both |
| Typical question | How much, how many, is there a difference | How and why, what is the experience | Both, one explaining the other |
| Tools | Instruments, structured questionnaires | Interviews, observation notes, documents | A combination |
| Analysis | Statistics | Coding into themes | Each kind analysed, then compared |
| Strength | Measures size and tests differences | Explains meaning and context | Gives both extent and reasons |
| Limit | May miss reasons and context | Findings are hard to generalise | Needs more time and skill |
Most laboratory measurement studies are quantitative. A qualitative strand is added when the question concerns people, such as why a safety procedure is not followed.
Deductive research starts with a theory or hypothesis and tests it with data. If theory says that a reaction rate rises with temperature, an experiment at several temperatures tests that prediction. Inductive research starts with observations and builds a general explanation from the pattern. Interviews that show the same obstacle to waste segregation in several laboratories lead to a general explanation. Many studies use both: observations suggest a hypothesis and an experiment tests it.
The main designs are introduced here, and the design of a particular study is worked out later in the unit.
| Design | What it does | Laboratory example |
|---|---|---|
| Experimental | The researcher changes one factor and measures its effect, with controls and replicates | Comparing two filter media on turbidity removal under the same conditions |
| Quasi-experimental | Compares groups without full random assignment | Comparing water from two sites that differ in treatment |
| Descriptive survey | Measures the characteristics of a group at one time | Measuring residue levels in samples collected from several sources |
| Correlational | Examines whether and how strongly two variables are related, without changing either | Relating turbidity to coliform counts |
| Case study | Examines one case in depth | Tracing the causes of a repeated contamination in one laboratory |
| Cross-sectional | Takes one measurement of many units at one time | One-day sampling of many boreholes |
| Longitudinal | Follows the same units over time | Measuring a river's dissolved oxygen every month for a year |
Example: a technician proposes "To determine whether heating raw honey above a set temperature increases its hydroxymethylfurfural content". The technician changes temperature and measures content, so the design is experimental. The aim is to explain a cause, so it is explanatory research with a quantitative approach and a deductive reasoning from the idea that heat drives the reaction.
Example: a technician proposes "To describe how laboratory staff record temperature of storage refrigerators and the reasons they give for missing entries". The data are practices and reasons, so the approach is qualitative with a descriptive purpose, using interviews and observation.
Create a free account to open more of this chapter.
Free: practical guides, quick cards, workplace scenarios and more.
Create a free account