2. A Few Keywords and Definitions for Understanding Statistics – IV (contd.)

 - by Dr. Prafulla Dikshit       (6 minutes read)

2.8  Types of Measures and tools

In the last post, we talked about the four levels of measurement of the data we collect for our research purposes. However, there is an entirely different aspect of data collection, which is the type of measure or the way or means through which data is collected or measured. Such measures of data may typically include tools or tests used to measure, collect, or gauge data. Such tools or tests normally vary with the field of study, for example, in psychology, we may have several standard tests measuring personality traits, and the intelligence quotient, for certain questionnaires focusing on a certain aspect of personality. In the medical field, professionals, use instruments like a BP instrument to get blood pressure, a thermometer to measure the temperature, or an electrocardiogram (ECG) to measure several vital statistics of the heart, such as the heart rate, the rhythm of the beats, and the strength and timing of the electrical impulses in the body. Similarly in mechanical engineering or metrology, instruments like a vernier caliper or a vernier height gauge are used to accurately measure linear distances. An atomic clock is a highly accurate time-measuring instrument, which is regulated by the atomic or molecular vibrations of elements like cesium or compounds like ammonia.

A closer look at these tools and measures indicates a pattern. These different measures and tools can be classified into some common types such as self-reported, standardized, observational, or physical tools and measures, irrespective of the research domain (Figure 1).

Figure 1

Types of measures

 

Notably, one, more, or all of these categories may apply to a given research domain. Psychology, for example, is one field, which involves situations wherein all of these types of measures are applicable. Table 1 shows the types of measures especially applicable in Psychology. These include Self-Report measures, Tests, behavioral, and physical measures, with respective psychology-specific examples.

Let’s talk a little more about each category. Each measure is distinct from the other not just in terms of the method of measurement, but also, in the pros and cons of using each method. The Self-reported measures, as the name suggests, are reported by the study subjects or participants, and are typically employed in research that employs human beings as the participants or subjects of study. This is about how people report they think, act, or feel. There are different ways in which self-reported measures can be employed. For example,  measures can help enquire how often people do something and are known as behavioral self-report measures. When people are asked to report their opinion about something, these are known as cognitive self-report measures, for example, whether they think a particular party will win the poll. Then some measures ask people to report how they feel about something, called affective self-report measures, such as questionnaires collecting data on how happy or contented, they are.

The Standardized measures are questionnaire instruments or scales that have been extensively tested prior to using them in real research, thus are relatively much more objective and non-biased measures of data collection, than the self-reported measures. For example, Aptitude tests, or IQ tests are standardized measures of the potential and intelligence of individuals respectively.

There are many situations when such standardization is either not possible or not desirable or both. For example, when we want to study an entire behavioral pattern or behavior over time, rather than a specific trait, we would want a measure and method that is more flexible. Observational measures are such methods or tools, which allow us to observe people, phenomena, or processes, as a whole. The observational measures involve making categorical or quantitative judgments about behaviors or phenomena of interest, by trained observers. For example observing how many cars stop at the signal, observing the flight patterns of birds, or for instance, whether and what proportion of people from an observed population show signs of a specific medical condition. Usually, such measures use some form of coding for the observations, which can then be processed for meaningful insights. 

Some measures are not directly observable nor can they be measured through psychographic or intangible tools from people, like questionnaires. Such measures are physical measures and usually require a physically and scientifically designed tangible tool for measurement. For example, it is not possible to directly observe blood pressure, but it can be measured through a B.P. instrument. Notably, these physical measures are much more objective than observational measures. These are not just observations of something happening, but a measure of a physical activity taking place in a system. When compared with physical measures,  standardized measures like questionnaires and tests may be considered almost at par, if not less in the objectivity of measurement. However, test questionnaires are different in terms of the tangibility of the instrument and the directness of the measurement. Physical measures are more direct than tests, and thus are likely to be less biased. However, physical measures are also not free from problems, as they involve human-being in their operation, and to that extent, they are subject to errors, bias, and subjectivity.  

 

Table 1

Examples of Types of Measures in the field of Psychology.

 

Insight 1 -> The above measures whether in general (Figure 1) or in a specific field like psychology can be seen as subjective versus objective. For example, the self-reported measures are more subjective than the standardized tests while the opposite is true for objectivity. Overall, if we arrange the measures in Table 1 in the order of increasing objectivity, it will look something like this:

Self-reported > Behavioral Measures > Tests > Physical Measures  

Insight 2 -> Even when we use objective measures such as a B.P. instrument cuff (a physical measure) for measuring the blood pressure, we may encounter measurement errors owing to various factors such as human errors in reading the Blood pressure levels, or owing to a faulty method of cuffing while measuring the B.P.

                                                                          ***