physiological data are typically accessible in most healthcare settings, allowing
for ease of data collection with minimal or no cost to the researcher. Researchers
must specify measurement protocols that include specific equipment used for
measurement, frequency and methods for calibration of equipment, training
for data collectors, specific times to obtain measurements, procedures for mea-
suring and recording data, and any special storage or handling considerations
(Merchant & Mateo, 1993).Issues in Quantitative Data Collection
Regardless of how the data are collected, researchers must consider a variety
of issues to ensure a high-quality study. They must have a written plan that
outlines the process for data collection, particularly when additional data col-
lectors are employed. Research assistants must be trained to collect data in a
very consistent manner (i.e., instruments should be administered in the same
order for all subjects, in the same context and setting, using the same set of
directions). Interrater reliability must also be established when more than one
person is involved in making observations. Interrater reliability is the extent to
which two or more individual raters agree. Although it is necessary for multiple
observers or raters to achieve interrater reliability, this may slow the process of
observational methods. Interrater reliability should be monitored periodically
throughout the study to increase the degree of confidence in the data (Casey,
2006). Researchers may choose to prepare a code book to organize the raw
data prior to collection and to assign numerical values to data obtained during
data collection. For example, researchers may assign a numerical value of 1 to
females and 2 to males (Wolf, 2003).
Data collection plans should detail a time frame. It is not uncommon for
researchers to encounter unplanned obstacles when gathering data. Frequently
data collection requires at least twice as long as the researcher anticipates.
Issues such as slow enrollment of consented subjects, heavy workloads, and
staff turnover are common causes of delay. Plans should include strategies to
manage attrition of subjects as a result of death, dropout, or relocation. Plans
also must address decisions about missing data. For example, subjects may fail
or refuse to respond to particular questions.
Many studies are funded by federal grants, state grants, or private foundation
monies. A budget is necessary to consider all of the factors involved in data
collection, including any delays that may be anticipated. Time extensions may
be requested by researchers because of delay, and this subsequently requires
researchers to account for any changes in budgeting. If the project is cut back
as a result of delays or exceeding the budget, the research can be seriously
compromised (Anastasi, Capili, Kim, & Chung, 2005).KEY TERMS
nominal: The
lowest level of
measurement
whereby data
are categorized
simply into groups;
categorical data
ordinal: A
continuum of
numeric values
where the intervals
are not meant to be
equal
interval: A
continuum of
numeric values
with equal intervals
that lacks an
absolute zero260 CHAPTER 10 Collecting Evidence