2.4 Types of RQs

All RQs have a population (P) and an outcome (O). However, different types of RQ emerge depending on whether the RQ also has an comparison/connection (C) or intervention (I).

This section studies different types of research questions:

These are compared in Sect. 2.4.4.

2.4.1 Descriptive RQs (PO)

Descriptive RQs are the most basic RQs, and identify:

  • The Population to be studied.
  • The Outcome of interest about this population.

Typically, descriptive RQs look like this:

Among {the population}, what is {the outcome}?

This is not a ‘recipe,’ but a guideline.

Example 2.12 (Descriptive RQ) Consider this RQ:

Among Australian males between 18 and 35 years of age, what is the average heart rate?

In this RQ, the Population is ‘Australians males between 18 and 35 years of age,’ and the Outcome is ‘Average heart rate.’ Notice that the Outcome is a numerical summary of the Outcome across the population (the average heart rate).

This is a descriptive RQ, as the RQ does not imply studying a connection with, or comparison between, the average heart rate and anything else.

Think 2.3 (POCI) Consider this RQ:

Among Australian adults, what proportion are coeliacs?

For this RQ, identify the Population and the Outcome.

Population: ‘Australian adults.’ Outcome: ‘the proportion that are coeliacs.’ (Note the proportion describes the population in general.)

2.4.2 Relational RQs (POC)

Usually, relationships are more interesting than just descriptions; relational RQs explore existing relationships. Relational RQs identify:

  • The Population.
  • The Outcome.
  • The Comparison or Connection.

Relational RQs have no intervention; the connection or comparison is not imposed by the researchers.

Typically, relational RQs based on a comparison look like this:

Among {the population}, is {the outcome} the same for {the groups being compared}?

Example 2.13 (Relational RQ) Consider this RQ:

Among Australians between 18 and 35 years of age, is the average heart rate the same for females and males?

In this RQ, the Population is ‘Australians between 18 and 35 years of age,’ the Outcome is ‘average heart rate,’ and the Comparison is ‘between females and males.’

This is a relational RQ based on a comparison. Notice that the average heart rate (Outcome) is a numerical summary across the two population sub-groups being compared (females; males).

The sex of the individual (the C) is not allocated by the researchers, so there is no intervention.

Typically, relational RQs based on a connection look like this:

Among {the population}, is {the outcome} related to {something else}?

Example 2.14 (Relational RQ) Consider this RQ:

Among Australians between 18 and 35 years of age, is the average heart rate related to age?

In this RQ, the Population is ‘Australians between 18 and 35 years of age,’ the Outcome is ‘average heart rate,’ and the Connection is with ‘age.’

This is a relational RQ based on a connection. Age (the C) is not allocated by the researchers, so there is no intervention.

Think 2.4 (POCI) Consider this RQ (based on Brown et al. (2000)):

In the Queensland Ambulance Service last year, what was the difference between the average response time to emergency calls between weekdays and weekends?

Identify the Population, the Outcome, and the Comparison.
Population: ‘Calls to the Queensland Ambulance Service’; the Outcome and the Comparison information come from these. Outcome: ‘average response time.’ Comparison: ‘between weekdays and weekends.’

Think 2.5 (POCI) Consider this RQ (based on Maron (2007)):

In Queensland state forests, is there a relationship between the average number of noisy miners and the number of eucalypts, in general?

(A noisy miner is a type of bird.) In this RQ, identify the Population. the Outcome, and the Connection.
Population: ‘Queensland state forests.’Outcome: ‘the number of noisy miners.’ Connection: ‘the number of eucalypts.’ Since the number of eucalypts would probably influence the number of birds, the Outcome would be the number of noisy miners.

Example 2.15 (Descriptive and relational RQs) Consider a study of blood pressure in Australians (the Population), comparing right- and left-arm blood pressures.

This is a descriptive RQ. There is no comparison, since there are not two subsets of the population being compared.

The blood pressure is measured twice on each member of the population: every member of the population is treated in the same way. The outcome is ‘the average difference between right- and left-arm blood pressure.’ This is a descriptive RQ.

In contrast, a study comparing the average blood pressure between females and males is a relational RQ. There is a comparison: the two subsets of the population (Australians) being compared are females and males.

2.4.3 Interventional RQs (POCI)

Interventional RQs explore relationships where the comparison/connection is determined or allocated by the researchers. They identify:

  • The Population.
  • The Outcome.
  • The Comparison or Connection.
  • The Intervention.

Interventional RQs may look like relational RQs, except that the comparison or connection is determined or allocated (i.e. imposed) by the researchers.

Sometimes it is not clear if the comparison or connection has been imposed by the researchers in an interventional RQ. When writing interventional RQs, make efforts to make it clear, if possible, when the RQ is interventional.

Example 2.16 (Interventional RQ) Consider this RQ:

Among Australians between 18 and 35 years of age, is the average heart rate for people allocated to receive a new pill the same as for people allocated to receive an existing pill?

In this RQ, the Population is ‘Australians between 18 and 35 years of age,’ the Outcome is ‘average heart rate,’ and the Comparison is ‘between those taking the new pill, and those taking the existing pill.’

There is an Intervention: the researchers allocate one of the pills to each subject. This is an interventional RQ.

Consider this RQ (McLinn et al. 1994):

In children with acute otitis media, what is the difference in the average duration of symptoms when treated with cefuroxime compared to amoxicillin?

In this RQ, the Population is ‘children with acute atitis media,’ the Outcome is ‘average duration of symptoms,’ and the Connection is between the types of drug (comparing ‘cefuroxime’ and ‘amoxicillin’).

It is not clear if there is an Intervention.

If the drugs are given to the children by the researchers, there is an intervention (giving the drug).

If the researchers just find children who are already taking the two drugs and measure the outcome (‘average duration of symptoms’), there is no intervention.

It is probable that there is an intervention.

2.4.4 Comparing the three levels of RQs

Descriptive RQs are the most basic and are usually used when a research topic is in its infancy; descriptive RQs set the platform for asking relational questions.

Relational RQs explore relationships, and provide an understanding of how the outcome of interest is related to certain sub-groups of the population; they may set the platform for asking interventional questions.

Interventional RQs (when possible to answer) are the most interesting: they can be used to test theories or models, or to establish cause-and-effect relationships (Table 2.1).

TABLE 2.1: The three types of RQs
RQ type P O C I
Descriptive (D) Yes Yes
Relational (R) Yes Yes Yes
Interventional (I) Yes Yes Yes Yes

Research often develops through these stages of RQs as knowledge grows and develops. For example:

  • Descriptive: What proportion of Australian adults are coeliacs (Cook et al. 2000)?
  • Relational: Among Australian adults, is the proportion of females who are coeliacs the same as the proportion of males who are coeliacs (Cook et al. 2000)?
  • Interventional: Among Australian adult coeliacs, is the percentage with adverse symptoms the same for those given a diet without oats and those given a diet with oats? (Janatuinen et al. 2002; Lundin et al. 2003)?

Think 2.6 (RQ types) What type of RQs are the following: Descriptive, Relational, or Interventional?

  1. Among Australian upper-limb amputees, is the percentage wearing prosthesis ‘all the time’ the same for transradial and transhumeral amputations? (Davidson 2002)
  2. In New York, what is the difference between the average height of oaks trees ten weeks after planting, comparing trees planted in a concrete sidewalk and a grassed sidewalk? (Grabosky and Bassuk 2016)
  3. What is the average response time of paramedics to emergency calls? (Pons et al. 2005)
  4. Is there a relationship between the average weekly hours of physical activity in children and the weekly maximum temperature? (Edwards et al. 2015)
1: Relational (using a Comparison). 2: Relational (using a Comparison) or Interventional, depending on whether the researchers plant the trees explicitly. 3: Descriptive. 4: Relational (using a Connection).