Personal tools

Academic writing templates

From UABgrid Documentation

Jump to: navigation, search


If you are writing a lit review then summarizing 100 articles for your own notes can be a daunting task. Having a guide helps ease the pain. Though not quite as stereotyped as, say, a haiku or a sonnet, many academic writing tasks have a recognizable structure. Since they have this common structure templates can be created. However, do not just "fill in the blanks". The templates are examples and suggestions and they should be altered as needed. After a few uses of a template you will not need it anymore. Below are template for use in the sciences. Choose and replace words, remove and add sentences as needed.

Empirical IMRD article summary template

adapted from Jennifer Greer’s Academic Paper Critique Template


In this article, Ego & Id (1892) explore/describe/identify research problem. They cite/propose prior research as showing background/context. The purpose of their study is to purpose. To that end, they postulate/ask hypothesis (or main question). To test their hypothesis/answer their question, Ego & Id selected sample population. They collected the following data using method. They found that main finding. This finding has broad/narrow implications in that implication. One important limitation of this study/paper is limitation. Nevertheless, this paper makes a contribution to the field with its study design/methodology/statistical rigor/bold or balanced point of view. The authors point to further areas of research as: future research.

example summary

Gouttard, S. et al., 2008. Assessment of Reliability of Multi-site Neuroimaging via Traveling Phantom Study. Medical Image Computing and Computer-Assisted Intervention–MICCAI 2008, 263–270.

In this article, Gouttard et al. (2008) investigate the reliability of multi-site neuroimaging studies. They propose that large scale multi-site neuroimaging studies like ADNI and ACE need special calibration and standardization protocols. The purpose of their study is to examine the reproducibility of neuroimaging results (they do not address accuracy as that is a question at different level of design). They propose that reliability can be achieved in multisite studies.

To answer thier question, Gouttard et al. scanned humans at four sites; three with Siemens Trios and one with an Allegra. They collected two repeated scans at each site. MPRage and T2 (TSE) are discussed in the paper. They characterized geometric distortions in raw phantom data and examined reliability of automated tissue segmentation. Sub-cortical structure segmentation was also examined.

They found that inter scanner variability was lower than intrascanner variability and that in fact a different scanner can produce distinctly different results. This finding has broad implications in that comparison of results from studies performed on different scanners may be difficult to compare.

Two important limitations of this study are that it depends on custom tissue segmentation algorithms and a small set of scanners. Nevertheless, this paper makes a contribution to neuroimaging with its methodolgy of using human phantoms a analyzing the joint variability of scanning and analysis and by providing variability estimates for future studies.