The effectiveness of your data-driven decision making is highly dependent upon the validity and reliability of the data from which those decisions are based.  To maximize the quality of your data, we can select the most appropriate research design for your project, implement statistical power analysis, develop sampling designs and procedures, design surveys & questionnaires, select or construct measures, scales or indexes, conduct factor analysis, and reliability & validity assessments.  Each of these services is described in more detail below.

Research Design.  After considering your study objectives, resource constraints, and ethical considerations, we can help determine the most appropriate research design for your project.  These could include cross-sectional, pre-experimental, experimental (e.g., classical, double-blind, Solomon four-group, Latin Square designs, fractional, factorial), and quasi-experimental designs (e.g., time-series, multiple time-series, nonequivalent control group designs).  We can also provide guidance and strategies for designing effective longitudinal studies (trend, cohort, and panel studies), and nested designs (e.g., cluster or group-randomized trials). 

Statistical Power Analysis.  Usually done at the early stages of a research project, statistical power analysis allows for the determination of the sample size that is most appropriate for your study i.e., so the study will be “sufficiently powered” for valid statistical tests.  Decisions about sample size are also usually made while also considering the level of precision you desire and your available resources.  Larger samples usually have more power than smaller ones.  Oversampling would lead to a waste of resources, as would under-sampling because, with insufficient power, a significant association or difference can go undetected.

Sampling Frames and Methods.  Ideally samples are representative of populations from which they are drawn.  Representative samples enable the generalization of results to the study population.  While probability sampling methods provide ways of deriving representative samples, there is always some degree of sampling error (i.e., samples will never provide perfect representations of populations).  Common sampling techniques include simple random, systematic, stratified, and multistage cluster sampling.  Though easier and cheaper, non-probability sampling methods (e.g., purposive and quota sampling) are less reliable and thus not recommended.

Survey and Questionnaire Design.  Interesting and engaging surveys, that are easy to complete, result in high quality data. We can reach respondents through a variety of channels, including online and mobile, in person (e.g., face-to-face interviews, paper-and-pencil, ACASI), mailed, or by phone, as well as in a variety of languages.  Multimodal research can help enhance response rates and data collection efficiency, help to maintain consistent sampling across modes, and improve data quality through the limitation of missing data.  Our experience with mailed surveys includes effective cover letters and questionnaire instructions, and various techniques to maximize response rates. We can assist with overall survey design and questionnaire construction, which includes the proper structure & formatting of questions (rating, matrix, semantic differentials, rankings, etc.), the proper sequencing of questions, as well as guidance to avoid questionnaire bias.  We can advise question wording, when closed or open-ended questions, or contingency questions are preferable, and provide guidance to avoid response set bias, and leading, threatening, or double-barreled questions.

Measurement/Scale and Index Construction.  When valid & reliable pre-existing measures/scales/indices are not available or suitable for your purposes, we can assist to construct new and innovative composite measures.  In doing so, we can help select the most appropriate scaling procedures (e.g., Bogardus social distance, Thurstone, Likert, semantic differentials, and Guttman scaling), and conduct validity and reliability assessments.  We can also assist with typology matrix construction and analysis, which is needed when researchers need to summarize the intersection of two or more variables, and results in the creation of sets of nominal categories.

Factor Analysis.  Factor analysis is applied to identify relatively few unobserved variables called “factors” by describing the variability among larger set of observed, correlated variables.  Whether performing exploratory factors analysis (EFA) or confirmatory factor analysis (CFA), the aim is to detect dimensions among a large number of interrelated variables.  EFA is employed when there are no a priori expectations about the number of factor, and how any indicator may be associated with any “factor”.  With CFA, hypotheses are tested about the number of factors and associated indicators in order to confirm whether they support theoretical assumptions.