The aim of the present research, which has taken a meta-analytical approach, has been to analyze and combine the analyses provided on the factors which influence test anxiety. The statistical population of the research consisted of all articles graded as scientific research, research projects conducted at research centers affiliated by the Ministry of Education and the thesis available at Razi and Mohaghegh Ardabili Universities through the years 2001 to 2015. Among the research studies conducted, 55 cases were selected for further examination. Data analysis had been done using CMA2, Excel and SPSS 22.0 software. Regarding descriptive statistics, frequency and percentage frequency were used, while in regard to inferential statistics, hexadecimal g and the fixed-effect and the random-effect model were used in order to measure influences, and the chi-squared test was used to examine the significance of the frequency, and the combination of Weiner T probability was used in order to combine the influence amounts. The I2 and Q tests were used in order to determine the homogeneity of the studies made, the results of which pointed out the preliminary studies to be not so. The calculated amounts and effects for the studies for both models -- fixed and random -- proved to be significant for both variables statistically. Furthermore, the results of having combined Weiner's t probabilities proved the combined effects to be significantly different. In whole, the study on factors which lead to test anxiety have resulted in 47 factors being identified as exerting large influences, and the factors of target orientation, training on attention control and meta-cognitive therapy have proved to have the highest influence upon test anxiety, a correlation which requires more attention on behalf of those in charge to the variables mentioned.