Preparing Needs Assessment Data for Analysis

¶ … Tailor data analysis to organizational and employee needs. Quantitative Data Analysis (Cohen & Crabtree, 2006) Quantitative data is tabulated so according to the results for different variables in the data set. This provides a comprehensive picture of the data and assists in the process of identifying patterns. A common way of displaying the data to facilitate the analysis is by using a frequency distribution, which is an organized tabulation of the number of the responses or scores according to each variable category.

Tabulation provides a structured way to determine data accuracy, identify data outliers, gauge the spread of the scores or responses, and observe categorical frequency Quantitative data analysis using interval data that are continuous, that has a logical order with standardized differences between values, but that does not have a natural zero. Items on a Likert scale are a good example of interval data. Data are frequently displayed in a manner that condenses the data from the constructed frequency and percent distributions.

Surveys: Closed-ended Questions Qualitative Content Analysis When content analysis is treated as a quantitative method of analysis, it provides a way to systematically and objectively analyzes media content (Rubin & Rubin, 2004). This version of content analysis used standardized measurements to code, characterize and compare texts (Warren & Karner, 2005).

When a qualitative approach is taken to content analytical, the focus is on analyzing both the explicit or manifest content of a text as well as interpreting the latent meaning of texts that which can be interpolated from the text, but that is not explicitly stated in it (Graneheim & Lundman, 2004).

The emphasis of content analysis is data coding, which may explain a major limitation of this approach -- its inability to provide a rich understanding of the meanings of texts In healthcare research, texts appropriate for content analysis include grant proposals, published manuscripts, minutes from meetings, transcripts of conversations, medical encounters, interviews, and focus groups (Cohen & Crabtree, 2006). Appropriate texts for analysis in healthcare fields also include messages communicated to the masses via newspapers, magazines, radio, television and the internet (Cohen & Crabtree, 2006).

Data can be presented in tables and matrices. This is useful particularly when quotations are used to articulate the findings by interweaving. What this means is that refinement of the analysis may well occur even as the manuscript is still being written in final form. Interviews Constant Comparative Method This qualitative data analysis method is a structured iterative process in which researchers compare each new bit of data with data that has already been examined in the study (Glaser & Strauss, 1967).

Open Coding: Each data bit is coded and then assigned to a relevant topic category or discarded if no relevance is observed. This coding occurs in accordance with how data bits compare with the accumulating corpus of analyzed data (Strauss & Corbin, 1990). Axial Coding: As the data bits are analyzed, new overarching topic categories will emerge. Once the data has all been coded and assigned to topic categories, the researcher examines the categories for emerging themes.

Theoretical saturation occurs when no new data appears to be emerging from the examined data (Strauss & Corbin, 1990). Selective Coding: In this last coding stage, the topic categories and the categorical interrelationships are used to create a storyline that tells or explains the phenomenon that is the focus of the research (Strauss & Corbin, 1990). The constant comparative method of data analysis can be used with structured responses, such as closed-end survey questions, or unstructured responses, such as those obtained when survey participants answer the open-ended items on a questionnaire.

That said, a constant comparative data analysis process has perhaps the most utility when used with extensive accounts that consist of unstructured data, such as interview transcripts. The presentation of findings in a constant comparative data analysis process is focused on revealing the themes that have emerged from the data. While visual displays of data may be used, the findings are.